Human-Annotated Benchmark

Vision-based driver monitoring systems are increasingly deployed in safety-critical intelligent transportation settings, yet they are almost always compared on classification accuracy alone. This paper argues that accuracy is insufficient to characterize a model's fitness for real-world deployment, and proposes the Human-Centered Benchmarking Framework (HCBF), which evaluates models across four dimensions: accuracy, explainability, efficiency, and robustness. The framework is applied to four representative lightweight architectures, MobileNetV3, ShuffleNetV2, EfficientNet-B0, and DeiT-Tiny, on the MRL Eye Dataset for eye-state classification. While the models are nearly indistinguishable on clean-set accuracy, each leads in exactly one dimension, and all four lie on the Pareto frontier. A Human-Centered Score computed under three deployment-oriented weighting scenarios ranks ShuffleNetV2 first throughout. However, this aggregate winner retains less than half of its performance under sensor noise and fails by classifying closed eyes as open, whereas the transformer remains robust. These findings show that aggregate ranking can mask dimension-specific vulnerabilities that are operationally decisive, underscoring the value of multi-dimensional, human-centered evaluation.

Multimodal Large Language Models (MLLMs) enable flexible instruction-driven image editing, but privacy risks arise when user images expose diverse and user-specific private content. Canonical privacy protection strategies typically substitute sensitive regions with surrogate content before cloud editing. Yet, the resulting output is often an edited surrogate rather than the desired edited source image, neglecting the local recovery in both design and evaluation scope. To this end, we introduce SPPE (Surrogate-based Privacy-Preserving Editing), the first recovery-oriented benchmark covering 36 fine-grained privacy categories and 65 editing instructions. It defines two complementary tasks: 1) editability assessment, which estimates before cloud interaction whether a surrogate can induce an edit consistent with the original image; and 2) surrogate-to-source edit recovery, which evaluates whether the edited surrogate can be transferred back to the private source with the edit effect preserved. We address each task with a dedicated method: ERMA predicts surrogate editability through instruction-aware multimodal relation modeling, while \method performs cycle-consistent recovery by using the surrogate editing pair as visual edit evidence and the source image as a source-preserving anchor. Experiments on SPPE and InstructPix2Pix show consistent improvements on both tasks. For editability assessment, ERMA improves over the best-performing baselines by 13.9% in SRCC and 12.3% in PLCC. For surrogate-to-source edit recovery, C2E-S2SER outperforms SOER across all 8 source integrity and edit consistency metrics on SPPE.

Recent advancements have significantly expanded the theoretical context windows of Multimodal Embedding Models (MEMs). However, larger context windows do not necessarily translate into effective comprehension and representation of long-context multimodal inputs, which remains a critical bottleneck for real-world deployment. To address the lack of systematic evaluation in this setting, we introduce MMLongEmbed, the first comprehensive benchmark for evaluating MEMs in long-context scenarios. MMLongEmbed comprises four retrieval tasks spanning multiple context-length ranges, covering text, document, and video modalities. Through extensive evaluation of state-of-the-art models, we find that current architectures rely heavily on superficial feature matching and struggle to capture deep semantic and structural dependencies. We further observe that performance degradation varies systematically with context length and key information placement. Moreover, models exhibit substantially different robustness to redundant contextual information across modalities. For reproducibility, the benchmark and code are publicly available.

While generative AI models have demonstrated remarkable success in structure-based drug design, they predominantly rely on deep binding pockets and struggle to sample effective ligands for challenging low-pocketability targets, such as the historically "undruggable" oncology targets KRAS and MYC. To address this gap, we introduce ShallowBench, a strictly curated benchmark of 5,780 shallow-pocket targets extracted from CrossDocked2020. By computing the difference between an Alpha Shape "lid" volume and the underlying protein atom voxel volume, we successfully isolated targets with low concavity while ensuring sufficient surface area for binding. Evaluating various state-of-the-art generative models reveals weaker predicted binding affinity on these low-concavity interfaces. ShallowBench therefore provides a rigorous benchmark for generative biology models and highlights the necessity of new architectural innovations or loss functions capable of navigating these challenging targets.

Recent advances in generative image editing have improved the realism and controllability of localized image manipulation, raising new challenges for image manipulation detection and localization (IMDL). However, existing IMDL benchmarks still have limitations in visual realism, manipulation diversity, and generator coverage, making it difficult to reflect recent trends in image manipulation. To address these limitations, we introduce Impostor, a high-quality AI-edited image manipulation localization dataset containing 100K manipulated images. Impostor is constructed by CraftAgent, a closed-loop agent framework that integrates scene perception, editing planning, manipulation execution, quality validation, and iterative reflection to automatically generate diverse and visually realistic manipulated images. Moreover, Impostor contains images generated by seven recent AIGC models across three manipulation types and includes multiple manipulated regions, providing a more comprehensive benchmark for AIGC-based IMDL. Furthermore, we propose PhaseAware-Net (PANet), a semantic-forensic framework that introduces local phase modeling and semantic-forensic consistency learning to better localize semantically plausible yet forensically disrupted manipulated regions. Extensive experiments show that Impostor poses significant challenges to existing large vision-language models (LVLMs) and specialized IMDL methods, while PANet achieves superior performance on Impostor and multiple public benchmarks.

Automated subject cataloging assigns controlledvocabulary headings to bibliographic records, but LCSH has no standard public benchmark. We introduce LCSHBench: 22,346 books in 15 languages from the openly licensed Harvard, Columbia, and Princeton catalogs. Records enter only when at least two independent cataloging agencies assigned LCSH; we release per-catalog provenance plus union and unanimous answer views. A concordance study of 465,187 works cataloged by all three libraries shows why this design matters: libraries usually agree on the underlying topic (93.3% share a concept-level heading) but often differ in exact expression (39.4% have identical heading sets). LCSHBench therefore scores both exact and concept matches, with set and rank metrics broken down by language and heading type, across open-vocabulary generation and full-vocabulary retrieval. As a first demonstration, a low-rank fine-tune of a 300M on-device embedder improves cross-lingual retrieval and beats a 3,072-dimensional hosted embedder on development exact recall@200 (0.659 vs 0.623). The language panel shows the gain is not uniform, and held-out-test and end-to-end confirmation remain future work.

Multimodal large language models (MLLMs) are predominantly evaluated on free-form vision-language tasks such as visual question answering, captioning, and summarization. However, their practical use is rapidly expanding to more structured computer vision settings, where users prompt models to perform localization-centric tasks such as object detection, often within larger agentic or decision-making systems. Despite this shift, there is currently no standardized benchmark that systematically evaluates these capabilities at scale. In this work, we introduce the first comprehensive benchmark specifically designed to assess the promptable localization abilities of generalist MLLMs. Our benchmark spans four core task categories: object detection, referring expression detection, instance-level detection, and video-based detection. To enable consistent and fair evaluation, we develop a unified framework that standardizes inputs, enforces parsable bounding box outputs, and defines transparent evaluation protocols across tasks. Using this suite, we evaluate a diverse set of open-source and proprietary MLLMs, providing an in-depth analysis of their performance and limitations. Beyond accuracy, we examine models' ability to adhere to output format specifications, showing that current systems are highly sensitive to formatting constraints and often fail to generalize even to minor variations. Our results highlight both the strengths and shortcomings of state-of-the-art MLLMs in localization settings, and point toward important directions for improving multimodal model design and evaluation.

We introduce HighTide, an evolving AI-assisted benchmark suite. Specifically, the contributions are: (i) a diverse open-source suite spanning multiple design languages and technology nodes, (ii) Bazel-based incremental RTL-to-GDS compilation with remote caching, (iii) AI-assisted design curation through twelve agent skills covering the design lifecycle, flow optimization, tool reference, and meta-maintenance, backed by per-design decision logs that serve as long-term memory of tuning rationale across the suite, and (iv) an infrastructure with RTL compilation verification for stable releases. The suite is publicly available and designed to grow with the open-source hardware ecosystem.

Multimodal agents in robotics, AR, and autonomous driving must reason about places and layouts from continuous egocentric streams, often using evidence outside the current view. Existing benchmarks either evaluate offline over full videos or target events rather than spatial structure. We introduce OVO-S-Bench, a fully human-annotated benchmark for streaming spatial intelligence, comprising 1,680 questions over 348 source videos. Annotation involves 12 trained annotators, each also serving as a blind cross-reviewer, across roughly 804 person-hours of multi-round quality assurance. Each question carries a query timestamp and an evidence interval, and at evaluation, the model sees only the prefix preceding the query. Questions span four levels of increasing abstraction: instantaneous egocentric perception, spatiotemporal context tracking, spatial simulation and reasoning, and allocentric mapping. Across 38 proprietary and open-source MLLMs, Gemini-3.1-Pro trails human experts by 27 points, 59.2 vs. 86.6, with allocentric mapping as the dominant bottleneck. Notably, streaming and spatially fine-tuned MLLMs underperform their own backbones. We further find that chain-of-thought reasoning amplifies spatial errors when ungrounded in the stream. By exposing these limitations, OVO-S-Bench establishes a demanding testbed for next-generation streaming spatial MLLMs.

This paper constructs the first benchmark on semi-supervised multi-modal crowd counting. To lay the foundation for this unexplored task, we first formulate the semi-supervised multi-modal setting and a standardized protocol that specifies the labeled-unlabeled data partition across different labeled ratios. Next, to establish solid reference points, we carefully tailor a diverse set of representative baselines, including existing fully supervised multi-modal methods and semi-supervised single-modal methods. Then, we carefully evaluate their performance under our proposed benchmark. Codes and the data partition will be released on https://github.com/HenryCilence/Semi-supervised-Multimodal-Crowd-Counting.

Assessing the quality of time series (TS) data is fundamental yet inherently challenging due to the multifaceted nature of quality dimensions. Recently, large language models (LLMs) have emerged as a promising paradigm for TS quality assessment via pairwise comparison and per-dimension evaluation. However, existing approaches rely on manually predefined quality dimensions and purely text-based reasoning, leaving it unknown whether LLMs can identify truly relevant quality dimensions or perform grounded and quantitative quality comparisons. To investigate this, we construct TSQBench, a dedicated benchmark for evaluating LLMs on two progressive capabilities: (i) understanding and identifying relevant quality dimensions, and (ii) performing quality comparison under specific dimensions. Our analysis reveals that current LLMs consistently struggle with both dimension identification and evidence-grounded quality comparison. To address these limitations, we propose TSQAgent, a novel agentic reasoning framework for TS quality rating consisting of three collaborative roles: Perceiver for focused dimension selection, Inspector for dimension-wise quantitative analysis, and Adjudicator that aggregates and refines the final judgment. In particular, we introduce an agentic reasoning strategy that instills the ability to identify and prioritize the most relevant quality dimensions, and further propose an agent workflow equipped with external analytical tools to enable precise quantitative comparisons over selected dimensions. Experiments on both the proposed benchmark and eleven real-world datasets demonstrate that our framework not only substantially improves LLMs' capabilities in quality understanding and quantitative comparison but also effectively translates these improvements into better quality-aware data selection, leading to enhanced downstream performance and data efficiency.

Visual posters are a compact medium for communicating dense information, yet progress on automatic poster generation remains difficult to measure because existing evaluations are often restricted to paper-only inputs, narrow domains, or surface-level visual similarity. We introduce Any2Poster Bench, a benchmark for any-source poster generation that evaluates systems across eight input modalities--PDFs, URLs, PPTX, DOCX, Markdown, LaTeX, notebooks, and videos--and five content domains. Any2Poster Bench pairs each source with quiz-based probes of verbatim factual retention and interpretive understanding, together with VLM-based judgments of visual quality, layout, readability, content completeness, and logical flow, enabling reproducible assessment of both information fidelity and visual communication. To instantiate and validate this benchmark, we further present Any2Poster Agent, an end-to-end reference agent that parses heterogeneous sources, organizes salient content, plans poster layouts, renders posters, and iteratively refines them using visual feedback. On Any2Poster Bench, Any2Poster Agent achieves 87.25% average accuracy across input modalities and 87.28% across content domains. On PaperQuiz-style evaluation, where prior paper-to-poster agents are directly comparable, Any2Poster Agent improves over PosterAgent-4o from 51.06-51.33% to 72.58% overall accuracy and from 116-121 to 145.16 in density-augmented score. Together, Any2Poster Bench and Any2Poster Agent provide a reusable evaluation resource and a competitive baseline for studying multimodal, domain-general poster generation.

Search-and-rescue (SAR) requires embodied agents to explore unfamiliar environments under multimodal uncertainty, perform multi-stage interactions, and retrieve spatial memory over long horizons. Existing benchmarks typically evaluate these capabilities in isolation, leaving unclear how failures compound when they must be composed in realistic workflows. We introduce RescueBench, a photo-realistic diagnostic benchmark that instantiates SAR as a four-stage pipeline: multimodal exploration, target rescue, memory-guided return, and final handoff. By combining sequential task composition with stage-level evaluation, RescueBench enables analysis of how exploration and memory failures propagate through embodied rescue workflows. It contains five progressive difficulty levels that vary in environmental complexity, clue ambiguity, and spatial hierarchy, along with an automatic episode generation and annotation pipeline for scalable evaluation and training. We evaluate seven baselines, an oracle reference, and human players, showing that no baselines complete the full task at the greatest difficulty. Stage-level diagnosis identifies autonomous exploration as the dominant failure mode and spatial memory as a second, independent bottleneck, suggesting that these limitations are not resolved by current topological visual-language navigation or map-based methods. Code is available in https://github.com/wukui-muc/RescueBench

Measuring the diversity of creative outputs is central to evaluating post-training mode collapse, comparing decoding strategies, and quantifying creative behavior in both AI and human writing. We propose a new approach to measuring diversity using in-context learning, of which the ``Decan'' metric, $D_{Ca_n} = C \times a_n$, is the working instance we evaluate: a per-byte score read off the per-token log-probabilities of a base model $θ$ in a \emph{single forward pass} per permutation, with no embedding model, no reference corpus, and no human labels. This approach is grounded in information theory, makes use of language model in-context learning to detect a wide range of similarities between any number of inputs, and obviates the need to train a special-purpose model. The same pipeline scores AI samples and human-written response sets, with diversity treated as a property of (responses, prompt, scoring model). On Tevet and Berant's human-grounded McDiv benchmark, $D_{Ca_n}$ reaches OCA 0.846 on the McDiv prompt_gen set where it performs best, behind the strongest neural baseline reported in Tevet and Berant (SentBERT, 0.897). On the OLMo-2-7B post-training pipeline, $D_{Ca_n}$ drops monotonically across the base $\to$ SFT $\to$ DPO $\to$ RLVR stages, detecting the type of diversity loss that creative-writing applications care about.

Agent-runtime systems emit traces, ledgers, provenance graphs, policy logs, delegation tokens, cache events, and tool-firewall records, but those containers do not necessarily answer governance questions about a specific decision. DEMM-Bench is a cross-regime benchmark for agent-runtime governance-evidence sufficiency, grounded in the Decision Evidence Maturity Model (DEMM): it measures whether records across eight evidence regimes are sufficient to reconstruct decision-level properties rather than merely present. The benchmark normalizes the regimes through adapters, asks property questions over actor, authority, action, policy, decision basis, resource touch, lifecycle context, and verification strength, and applies eight deterministic degradation conditions. Across 64 manuscript cases, trace-present and schema-present baselines overclaim on 75% of cases, ledger-present overclaims on 50%, and the redacted property-level candidate scorer has zero overclaim with 56.25% mean Property Sufficiency Accuracy. The deposited package provides the 64-case dataset, construction-oracle labels, baselines, and adapters, supporting reproducible evaluation of decision-evidence maturity across heterogeneous agent-runtime evidence substrates.

Safe human--robot collaboration requires more than visual description: a monitor must determine whether the robot body is safely separated, already colliding with the scene or a person, or about to collide. We call this capability collision grounding: binding visual observations to robot body geometry, camera viewpoint, scene layout, human proximity, and temporal motion in order to infer present and imminent contact. We introduce TouchSafeBench, a physics-grounded benchmark for evaluating collision grounding in vision-language models (VLMs). Built in Habitat~3.0, TouchSafeBench contains 2,940 simulated indoor co-presence episodes across social navigation and social rearrangement, with synchronized multi-view RGB-D observations, top-down trajectory maps, calibrated camera metadata, and simulator-derived contact labels. We study two deployment-facing tasks: classifying the current safety state and warning about imminent collision before contact. Across three frontier or robotics-oriented VLMs and nine visual representations, current models remain far from reliable: the best average Macro-F1 stays below 50%, explicit depth is not automatically transformed into robot-body collision evidence, and robot--scene contact is consistently harder than human-contact risk. TouchSafeBench reveals a central limitation of embodied VLMs: visual fluency does not imply physical accountability. Reliable robot safety monitors will need representations that explicitly bind viewpoint, robot morphology, metric geometry, and future collision. We will release the benchmark upon acceptance.

We introduce Office Comprehension Bench (OCB), the first public benchmark to jointly evaluate LLM systems on Word, Excel, and PowerPoint comprehension over native file formats (.docx, .xlsx, .pptx) and their variants. OCB consists of two tracks. File Fidelity Q&A tests structural and visual perception of office artifacts - tables, charts, embedded images, formulas, and app-specific elements such as headers, speaker notes, and named ranges. Domain Q&A tests expert-level reasoning grounded in real-world industry documents across 12 professional domains, with queries requiring multi-step analysis and synthesis across documents. Each reference answer is decomposed into atomic, binary-gradable claims, and an ensemble of LLM judges scores responses against each claim independently. Even the strongest frontier system in its default reasoning mode reaches only about 59.3% on Domain Q&A; increasing thinking depth within a tier does not move performance materially, while moving to a higher product tier yields modest gains. We release the dataset, evaluation tooling, judge prompt, and a public leaderboard.

Classroom videos contain observable teaching practices, but their pedagogical and visual signals are rarely organized in forms suitable for model evaluation. We present \textit{TeachObs}, a human-validated benchmark for multimodal teaching observation in classroom videos. \textit{TeachObs} includes 30 public lesson videos from eight countries divided into 5,158 fixed 15-second scenes. Seven researchers annotated each scene with 39 binary observation codes, covering 20 visual codes, such as gesture, board work, pointing, and visual materials, and 19 nonvisual codes, such as instruction, monitoring, questioning, feedback, and reflection. Gold segment labels are constructed using reliability- and prevalence-aware rules based on Krippendorff's alpha. In addition to segment-level labels, three expert raters produced lesson-level ratings and qualitative evaluations of instructional design, instructional delivery, learner response, learning materials, and lesson closure across the 30 lessons, with rater coverage detailed in the body. Using these two human reference layers, we evaluate five vision-capable frontier LLMs across three tracks - text-only segment coding, text + frame segment coding, and lesson-level coverage scored under an LLM-as-judge protocol - and find that no single model consistently outperforms others across all three tracks, that adding a mid-frame inflates both true and false attributions per scene, and that model evaluations over-rate procedurally clear lessons relative to expert raters. \textit{TeachObs} therefore supports both fine-grained annotation benchmarking and whole-lesson evaluation, showing where AI systems can assist classroom video analysis and where expert judgment remains necessary across varied subjects, classroom formats, and annotation difficulty levels.

Real-world data analysis is inherently iterative, yet existing benchmarks mostly evaluate isolated or short interactive tasks, leaving agents' ability to track evolving analytical context over long horizons untested. We introduce LongDS, a benchmark for long-horizon, multi-turn data analysis where agents must maintain, update, restore, and compose evolving analytical states. LongDS comprises 68 tasks constructed from real-world Kaggle notebooks, spanning 2,225 turns across six domains including Geoscience, Business, and Education. Tasks are designed around state-evolution patterns (e.g., counterfactual perturbation, rollback, multi-state composition), with an average dependency span of 11.3 turns. Evaluating five state-of-the-art models, we find that the best model reaches only 48.45% average accuracy, performance drops nearly 47 points from early to late turns, and long-horizon errors account for 52%--69% of failures. Further analysis shows that additional agent steps do not necessarily improve performance, suggesting that the key bottleneck is maintaining a correct analytical state rather than increasing interaction budget. We release LongDS to support research on reliable long-horizon agentic data analysis. Code and data will be released at https://github.com/zjunlp/DataMind.

We benchmark three supervised fine-tuned models against frontier zero-shot baselines on a 661-row held-out slice of PiSAR (Persona, intent, Screen, Action, Rationale), a 12,929-tuple corpus of screen-anchored behavioural rationales curated from public app-store reviews, Pew American Trends Panel demographics, and the OPeRA shopper traces. Every model, frontier or fine-tuned, is evaluated on the same 661-row slice with the same scoring pipeline. Two findings. First, frontier zero-shot baselines (Claude Opus 4.7 and GPT-5.5) reach sem_sim 0.459 and 0.482 respectively; a fine-tuned Qwen3-VL-8B-Instruct reaches 0.783 and clears sem_sim >= 0.7 on 79% of rows, against 1-2% for either frontier baseline, a gap of 0.30 absolute on the same test set. Second, the same training data and recipe on Gemma-4-26B-A4B-IT scores only 0.441, in the same band as the frontier zero-shot baselines rather than the fine-tuned Qwen. We read this as a recipe-vs-model mismatch: the reasoning-tuned high-parameter model resists displacement and would likely need either more data or a stronger fine-tuning method.

AI model hubs provide access to a rapidly growing collection of powerful pre-trained models, enabling off-the-shelf mixture-of-experts systems with different routing strategies. However, this rapid growth poses two fundamental challenges: scaling model selection across thousands of experts and continually updating routing mechanisms as new models and tasks are introduced. In this paper, we formalise this setting as Continual Model Routing (CMR) and propose CMRBench, a new large-scale benchmark simulating realistic hub expansion and including over 2,000 candidate models. Finally, we introduce CARvE, a contrastive embedding approach for efficient continual model routing via checkpoint-based anchoring and structured replay. Extensive empirical results and ablations show that CARvE significantly outperforms zero-shot retrieval, fine-tuning, and adapter-merging baselines in model, family, and domain-level accuracy.

A coding agent executes a benign task as a sequence of shell, file, and network actions, any of which can quietly exceed the authorized scope while the task still completes. We call this overeager behavior: the prompt is not adversarial and the run succeeds, yet an out-of-scope step can leak credentials or delete files. Existing benchmarks miss it: task-completion suites credit any finished run, jailbreak suites probe adversarial prompts, and the one prior overeager benchmark applies a single fixed prompt set to every agent-model pair, leaving its easiest and most resistant pairs under-measured. We present SNARE (Synthesizing Non-adversarial scenarios for Adaptive Reward-guided Elicitation), a pipeline that composes benign scenarios from reusable scope and trap fragments, scores each run with a judge-free oracle flagging trap-pattern matches and unsolicited file additions or deletions, and uses Thompson sampling to steer each pair's run budget toward the scenarios that most often trigger it. Instantiating it over 24 overeager archetypes yields OverEager, which we run across a 4x5 matrix of four coding agents and five base models. Across 10,000 benign runs, 19.51% trigger overeager behavior, with per-pair rates spanning 11.9x. This variation is driven by the agent framework, not the model: the framework accounts for 56% of it against the model's 21%, so any single-framework or single-model evaluation undercounts the matrix by about a fifth.

Video generation models (VGMs) are rapidly entering classrooms, yet existing benchmarks evaluate only perceptual quality, intrinsic faithfulness, generic safety, or video as a reasoning medium, and none assesses whether the outputs are educationally valid. In this work, we present EduVideoBench, the first balanced benchmark in the education domain, grounded in the Knowledge-Skills-Attitude (KSA) framework so that pedagogical adequacy and educational safety are evaluated jointly rather than as ad-hoc quality dimensions. Across five frontier VGMs, our results show substantial room for improvement across knowledge, skills, and attitude before they are classroom-ready. We complement this with a qualitative analysis of expert comments, finding that educational validity is multi-component, where a single misaligned element such as pacing, legibility, or notation can invalidate an otherwise correct video. We hope EduVideoBench will guide the development of VGMs that are pedagogically grounded and safe for the classroom.

Large language models (LLMs) now support automated software security tasks, including vulnerability discovery and proof-of-concept (PoC) generation. Existing benchmarks do not faithfully evaluate LLMs in real-world bug hunting scenarios because they rely on fuzzing harnesses, target-specific descriptions, or vulnerability-reproduction tasks. We present SEC-bench Pro, a benchmark for measuring agent bug hunting on critical, high-complexity software systems. This work discloses reports with concrete PoC inputs and links fixes into reproducible tasks through a three-phase pipeline for vulnerability collection, environment reconstruction, and oracle-based validation. We instantiate SEC-bench Pro with 183 validated vulnerabilities across V8 and SpiderMonkey, including a V8 subset with more than $1.5 million in cumulative Google Vulnerability Reward Program awards. These instances span memory-safety, sandbox, JIT, and race-condition bugs under browser-grade and runtime-grade execution conditions. Our evaluation shows that coding agents with frontier models remain below 40% success on both evaluated engines. The open-weight Kimi-K2.6 baseline reaches 11.7% on V8, while the strongest frontier configuration reaches 32.0% on V8 and 38.8% on SpiderMonkey. ClaudeCode and Codex solve complementary instance sets, and their two-agent union reaches 37.9% on V8 and 48.8% on SpiderMonkey. SEC-bench Pro provides robust environments for assessing LLM-based security agents and exposes limitations in long-horizon bug hunting tasks.

AI agents are beginning to complete valuable, long-horizon business operations tasks, but training and evaluation environments for enterprise work still struggle to balance realism, verifiability, and scale. Environment and task creation frequently suffers from a failure mode we call artifact drift: when instructions, environments, oracles, and verifiers are created by loosely coupled processes, they frequently disagree on what a task requires, producing environments that are unsolvable, reward-hackable, or inconsistent. We introduce Anchor, a task-generation pipeline that formalizes domain experts' specifications of business workflows into constraint optimization programs. From a single parametric specification, the pipeline jointly produces a natural-language instruction, environment configuration, solver-certified ground-truth solution, and state-based verifier. With Anchor, altering parameters yields new tasks with controlled difficulty and known optimal solutions, producing harness-agnostic environments whose rewards depend solely on end-state business correctness. We apply Anchor to produce ERP-Bench: a benchmark of 300 long-horizon tasks spanning procurement and manufacturing workflows in a production-grade ERP system. We find that generation parameters predict realized difficulty, and that frontier models satisfy explicit task constraints in 26.1% of trials but reach a fully optimal solution in only 17.4% of trials. Overall, we show that Anchor and ERP-Bench offer a concrete recipe for building auditable evaluation environments for economically valuable agent work. We release the task generator and ERP-Bench dataset at erpbench.ai

Benchmarks increasingly guide deployment, procurement and scientific screening, yet a score supports only the response it records, not necessarily the deployment action. We introduce deployment-complete benchmarking, which tests whether benchmark evidence determines a deployment action. A benchmark is complete for a claim exactly when the action is constant on each evidence fiber; mixed fibers expose missing deployment information, and completion curves quantify the evidence required to resolve ambiguity. In controlled response spaces, benchmark-channel conformal coverage of 94.98% transferred poorly to an unmeasured deployment channel (10.07%), whereas response-rank intervals achieved 94.91% coverage; even zero benchmark error certified only 45.4% of candidates at the largest residual size. Public audits revealed incompleteness, including 97.9% mixed Tox21 fibers and zero median certifiable fraction in main Matbench and JARVIS audits. In held-out replays, certify-then-acquire reduced false decisions from 1.19% to 0.027% in Tox21 and from 20.3% to 0.128% in JARVIS, while changing model choice and identifying deployment-relevant probes. Deployment-ready benchmarks should report evidence, supported actions, ambiguity and completion cost rather than scores alone.

Accurate evaluation of weather forecasting models is critical for their reliable deployment in real-world applications. However, existing benchmarks predominantly rely on reanalysis products such as ERA5, which are generated through delayed data assimilation and do not reflect the constraints of real-time operational forecasting, thereby resulting in a systematic mismatch between benchmark performance and real-world forecasting. In this work, we introduce RealBench, a next-generation benchmark for AI weather forecasting that emphasizes realistic evaluation under operational conditions. RealBench features a strictly out-of-distribution test set spanning 2025 to eliminate data leakage and capture recent atmospheric regimes. It integrates multiple data sources, including low-latency operational analysis and a large-scale global in-situ observation dataset comprising over 10,000 stations, enabling direct evaluation against real atmospheric measurements. Beyond standard global metrics, RealBench provides a comprehensive evaluation framework for high-impact extreme events, including heatwaves, cold surges, and tropical cyclones, using event-specific metrics that better reflect real-world forecasting priorities. The evaluation results reveal substantial discrepancies between reanalysis-based metrics and real-world performance, particularly concerning extreme events. By highlighting the limitations of existing benchmarks, this work establishes a more faithful and operationally relevant evaluation paradigm, providing a rigorous foundation for advancing next-generation AI weather forecasting systems. The benchmark implementation is available at: https://github.com/lixruize-del/NWP-Benchmark.

While large language models (LLMs) hold transformative potential for medicine, their reasoning robustness and safety in real-world clinical scenarios remain critically underexplored, particularly in dentistry. Here we introduce GlobalDentBench, the first multinational dental benchmark, featuring a taxonomy that encompasses 14 dental specialties across 88 countries and regions spanning six continents. The benchmark comprises 8,978 expert-validated questions across three formats (multiple-choice, short-answer, and case-based questions) and assesses three progressive reasoning levels: knowledge recall (L1), routine reasoning (L2), and individualized reasoning (L3). To ensure data quality, the automated construction framework was calibrated by six senior dentists, achieving expert agreement rates of 99.98% for multiple-choice and short-answer questions and 96.78% for the more complex case-based questions. Evaluation of 12 frontier LLMs on GlobalDentBench revealed a sharp, stepwise performance degradation with increasing reasoning complexity. Specifically, accuracy plummeted from 81.34% on multiple-choice to 64.53% on short-answer and 22.34% on case-based questions, while declining markedly from 74.01% at L1 to 55.64% at L2 and 35.71% at L3. More critically, risk analysis of real-world dental cases demonstrated an alarming overall unsafe rate of 31.01% in LLM-generated clinical recommendations, with 4.51% posing risks of irreversible patient harm and risks particularly pronounced in specialties such as orthodontics. These findings expose fundamental limitations in the medical reasoning and safety of current LLMs. Consequently, GlobalDentBench provides a scalable foundation for trustworthy clinical AI evaluation, underscoring the urgent need for rigorous validation before the safe deployment of these models in healthcare.

Understanding why a spacecraft maneuvers -- rather than simply that it did -- is an increasingly important problem for space domain awareness as Earth orbits grow crowded and contested. Current analysis pipelines are built for detection: they are good at picking up that something happened, less good at reasoning about what it means. AstroMind is a physics-grounded benchmark designed to close that gap. It draws on high-fidelity astrodynamics simulations and real observational constraints, converting them into verifiable reasoning problems across three task types: intent inference, maneuver parameter estimation, and threat assessment. Each scenario includes realistic sensing noise and multi-source textual intelligence at varying reliability levels. Evaluation metrics capture both semantic correctness and quantitative consistency under physical constraints. Benchmarking a suite of open-weight models shows no single model dominates every axis: Qwen3 (32B) leads on intent inference accuracy; QwQ (32B) leads on threat assessment and achieves the lowest median relative error on parsed items; GPT-OSS (20B) produces the strongest judged reasoning quality and extracts the most scalar values for parameter estimation (136 of 241 parsed items). Training data composition and reasoning style matter as much as model size. Structured reasoning prompts help consistently across tested 8B models, with larger gains for those that can already track physical constraints. AstroMind gives the field a shared test for a problem where getting the physics right and reading the tactical situation correctly are both required -- neither is sufficient on its own.

Reinforcement learning (RL) has emerged as a viable recipe for training LLM agents to reason over external memory banks in multi-session dialogue. Existing work trains exclusively on a single benchmark, leaving open how the composition of training data shapes the skills a memory agent acquires. We present a controlled empirical study that holds architecture, RL algorithm, and all hyperparameters fixed and varies only the training curriculum across three conditions: in-domain (LoCoMo), mixed-benchmark (LoCoMo + LongMemEval), and out-of-domain (LongMemEval only). Across two benchmarks and ten question types, curriculum composition acts as a fine-grained lever on specialization rather than a uniform scaling factor on performance. The mixed curriculum yields the strongest overall F1 on both evaluation sets. Training on a narrow out-of-domain set transfers a targeted skill - temporal reasoning - despite weak aggregate performance. Per-type differences substantially exceed aggregate differences, indicating that single-number benchmark comparisons systematically underreport curriculum effects. We further report two practical lessons from adapting GRPO to a single-GPU regime: cross-benchmark mixing requires filtering format-specific noise from memory banks to preserve training signal, and binary exact-match reward produces no learning signal at the small group sizes (G = 4) required on one GPU, motivating continuous reward functions in this regime.

Extracting structured information from visual documents (Visual Information Extraction, VIE) is a cornerstone of business automation. While recent Multimodal Large Language Models (MLLMs) have shown promising capabilities, existing benchmarks suffer from critical limitations in scale and realism, lack semantic granularity, and fail to cover diverse document types. To bridge this gap, we introduce ReceiptBench, a large-scale, human-annotated benchmark consisting of 10k diverse receipts, organizing information extraction into four hierarchical sub-tasks: (1) Basic Perception for raw text spotting, (2) Format Normalization for strictly following standardization instructions, (3) Semantic Reasoning for inferring implicit attributes from context, and (4) Structure Parsing for handling nested line items. Furthermore, we propose a two-stage training framework incorporating Metric-Aware Group Relative Policy Optimization (GRPO), which translates rigorous evaluation constraints into reinforcement learning signals to enhance structural consistency. Extensive experiments demonstrate that our method yields state-of-the-art performance, surpassing leading proprietary models on complex reasoning tasks. We release our datasets and code at https://github.com/wwwT0ri/ReceiptBench.

Recent advances in large language models and tool-using agents have expanded the range of benchmarked web tasks. Yet an important class of specialized retrieval tasks remains undercharacterized. On many specialized data-retrieval websites, answer-bearing evidence becomes accessible only after establishing the correct site-specific retrieval state through filters, views, hierarchies, or scopes. We term this capability state-gated retrieval (SGR). We introduce SGR-Bench, a benchmark for this setting containing 100 expert-curated tasks spanning six source families and 12 public data ecosystems. Each task requires discovering the appropriate website and configuring its site-specific retrieval state to produce a structured answer. SGR-Bench pairs constraint-guided and goal-oriented formulations of the same underlying problems, enabling controlled comparisons between explicit and implicit guidance for state-gated retrieval. We evaluate eight CLI-based agentic LLM systems and three commercial search-agent products. On SGR-Bench, the strongest system reaches only 66.18% item-level F1, while row-level F1 remains much lower. A manual audit of 156 analyzable failed CLI trajectories shows why: agents often reach a relevant web source, but establish the wrong site-specific retrieval state. Retrieval-scope drift (37.2%) and criterion mismatch (27.6%) dominate, whereas final answer composition accounts for only 10.3%. The dataset and single-case evaluation instructions are available at https://huggingface.co/datasets/PKUAIWeb/SGR-BENCH.

Assessing progress toward the Sustainable Development Goals (SDGs) requires multi-step reasoning over visual cues, contextual knowledge, and development indicators, where incomplete evidence use and imperfect evidence integration can introduce hidden prediction biases. Real-world SDG monitoring further spans both qualitative judgments and quantitative estimation. However, existing benchmarks typically evaluate these aspects in isolation, obscuring systematic biases that emerge when models substitute priors for evidence. To address this gap, we propose SDGBiasBench, a large-scale benchmark suite for SDG-oriented vision-language reasoning. Spanning 500k expert-involved multiple-choice questions and 50k regression tasks, the benchmark enables comprehensive assessment of both decision-level and estimation-level bias in Vision--Language Models (VLMs). Evaluations on SDGBiasBench reveal an intrinsic SDG bias in current VLMs, where predictions are frequently driven by SDG specific priors rather than reliable multi-modal cues. To mitigate such bias, we propose CADE (Contrastive Adaptive Debias Ensemble), a training-free, plug-and-play method that leverages modality-specific answer priors. CADE yields significant gains on the proposed benchmark, improving multiple-choice accuracy by up to 25% and reducing regression MAE by up to 12 points across multiple VLMs. We hope our work can foster the development of more fair and reliable AI systems for sustainable development.

Vision-language models reliably name objects in a scene, but do they represent the 3D layout those objects inhabit? We introduce a 3,034-sample human-curated benchmark targeting three components of spatial understanding: depth-ordered occlusion (probed via three independent counterfactual operationalisations), optical-geometry inference over visible reflections, and volumetric rearrangement planning. Six frontier and open-weight VLMs, scored by trained annotators on 18,204 responses with no LLM-as-judge, reveal a sharp dissociation: models that plan rearrangements over visible layouts at 53--97% accuracy and rarely violate collision constraints fall to 6--45% on occlusion and below 7% on reflections. An embodied-reasoning model reproduces the same profile. White-box analysis on Qwen3-VL-8B-Thinking localises the failure to the visual-token merger: spatial information recoverable throughout the vision encoder becomes inaccessible after token compression and only stabilises again when clean post-merger activations are patched into the language decoder.

Video generation is rapidly evolving from single-shot synthesis to complex multi-shot audio-video (MSAV) narratives to meet real-world demands. However, evaluating such frontier models remains a fundamental challenge. Existing benchmarks are limited in scope and data diversity, and rely on rigid evaluation pipelines, preventing systematic and reliable assessment of modern MSAV models. To bridge these gaps, we introduce MSAVBench, the first comprehensive benchmark and adaptive hybrid evaluation framework for multi-shot audio-video generation. Our benchmark spans four key dimensions, video, audio, shot, and reference, covering diverse task settings, varying shot counts of up to 15, and challenging non-realistic scenarios. Our evaluation framework improves robustness through an adaptive self-correction mechanism for shot segmentation, instance-wise rubrics for subjective metrics, and tool-grounded evidence extraction for complex judgments. Furthermore, MSAVBench achieves high alignment with human judgments, reaching a Spearman rank correlation of 91.5%. Our systematic evaluation of 19 state-of-the-art closed- and open-source models shows that current systems still struggle with director-level control and fine-grained audio-visual synchronization, while modular or agentic generation pipelines offer a promising path toward narrowing the gap between open- and closed-source models. The benchmark data and evaluation code are publicly available at https://github.com/ali-vilab/MSAVBench.

Vision-Language Models (VLMs) have demonstrated remarkable capabilities in general video understanding, yet they often struggle with the fine-grained comprehension crucial for real-world applications requiring nuanced interpretation of human actions and interactions. While some recent human-centric benchmarks evaluate aspects of model behaviour such as fairness/ethics, emotion perception, and broader human-centric metrics, they do not combine long-form videos, very dense QA coverage, and frame-level spatial/temporal grounding at scale. To bridge this gap, we introduce FineBench, a human-centric video question answering (VQA) benchmark specifically designed to assess fine-grained understanding. FineBench comprises 199,420 multiple-choice QA pairs densely annotated across 64 long-form videos (15 minutes each), focusing on detailed person movement, person interaction, and object manipulation, including compositional actions. Our extensive evaluation reveals that while proprietary models like GPT-5 achieve respectable performance, current open-source VLMs significantly underperform, struggling particularly with spatial reasoning in multi-person scenes and distinguishing subtle differences in human movements and interactions. To address these identified weaknesses, we propose FineAgent, a modular framework that enhances VLMs by leveraging a Localizer and a Descriptor. Experiments show that FineAgent consistently improves the performance of various open VLMs on FineBench. FineBench provides a rigorous testbed for future research into fine-grained human-centric video understanding, while FineAgent offers a practical approach to enhance such reasoning in current VLMs. Project page and code at https://joslefaure.github.io/assets/html/finebench.html.

Large Language Model (LLM) agents are increasingly applied to engineering design tasks, yet existing evaluation frameworks do not adequately address multi-agent systems that combine simulation, retrieval, and manufacturing preparation. We introduce a benchmark suite with three evaluation dimensions: (1) a workflow benchmark with seven prompt styles targeting distinct cognitive demands-including direct tool use, semantic disambiguation, conditional branching, and working-memory tasks; (2) a Retrieval-Augmented Generation (RAG) benchmark with gated scoring isolating retrieval contributions to parameter selection; and (3) an High Performance Computing (HPC) benchmark evaluating end-to-end ML training orchestration on a SLURM cluster. Alongside the benchmark we present EngiAI, a Multi-Agent System (MAS) reference implementation built on LangGraph that operationalizes the benchmark by coordinating seven specialized agents through a supervisor architecture, unifying topology optimization, document retrieval, HPC job orchestration, and 3D printer control. Across four LLM backends and two EngiBench problems, proprietary models achieve 96-97% average task completion on Beams2D, while open-source 4B-parameter models reach 55-78%, with clear generational improvement. Conditional branching proves most challenging, with task completion dropping to 20-53% for the conditional style on Photonics2D. RAG gating confirms near-perfect retrieval-augmented scores (about 1.0) versus near-zero without retrieval, validating the evaluation design. On HPC orchestration, one model completes all pipeline steps in 100% of runs while another drops to 50%, revealing that multi-step instruction following degrades over long-running workflows.

CAPTCHAs are widely deployed as human verification mechanisms and frequently block intelligent agents from completing end-to-end automation in real-world web environments. Solving modern CAPTCHAs requires robust multi-step visual reasoning and interaction capabilities, yet training-based approaches have remained absent due to the lack of large-scale training data and process-level annotations. We introduce CaptchaBench, the first CAPTCHA benchmark designed to support large-scale training, comprising 16,000 programmatically generated samples across eight task categories with detailed region and process-level annotations. Systematic evaluation on CaptchaBench reveals that existing methods fail consistently on tasks requiring fine-grained visual detail capture and region-level comparison. We therefore present CaptchaMind, an RL-based solver trained with explicit reasoning process supervision, achieving 82.9% average success rate across eight tasks and 71.0% on real-world instances, substantially outperforming all existing methods without closed-source APIs.

Hallucination remains a central failure mode of large language models, but existing benchmarks operationalize it inconsistently across summarization, question answering, retrieval-augmented generation, and agentic interaction. This fragmentation makes it unclear whether a mitigation that works in one setting reduces hallucinations across contexts. Current benchmarks either require human annotation and fixed references that may be memorized, or rely on observations in settings that are difficult to reproduce. To study root causes, we introduce HalluWorld, an extensible benchmark grounded in an explicit reference-world formulation: a model hallucinates when it produces an observable claim that is false with respect to this world. Building on this view, we construct synthetic and semi-synthetic environments in which the reference world is fully specified, the model's view is controlled, and hallucination labels are generated automatically. HalluWorld spans gridworlds, chess, and realistic terminal tasks, enabling controlled variation of world complexity, observability, temporal change, and source-conflict policy, and disentangling hallucinations into fine-grained error categories. We evaluate frontier and open-weight language models across these settings and find consistent patterns: perceptual hallucination on directly observed information is near-solved for frontier models, while multi-step state tracking and causal forward simulation remain difficult and are not generally solved by extended thinking. In the terminal setting, models also struggle with when to abstain. The uneven profile of failures across probe types and domains suggests that hallucinations arise from distinct failure modes rather than a single capability. Our results suggest that controlled reference worlds offer a scalable and reproducible path toward measuring and reducing hallucinations in modern language models.

Existing benchmarks for LLM coding agents primarily evaluate final outcomes. While useful for measuring overall capability, these metrics provide limited visibility and often miss defects that arise during execution. We present ProcCtrlBench, a benchmark for execution-process evaluation in LLM coding agents. ProcCtrlBench organizes recurrent execution defects into a reusable ontology covering 11 defect types in 4 categories, and evaluates agent trajectories through standardized process evidence rather than final outcomes alone. To support comparison across heterogeneous agents, ProcCtrlBench standardizes raw logs into a unified trajectory representation and reports calibrated scorecards over process-level findings. In addition, ProcCtrlBench uses control preservation as a way to quantify execution-process quality, capturing whether execution remains interpretable, interruptible, correctable, reversible, and able to hand back authority when needed. We evaluate ProcCtrlBench on 200 cases sampled from three benchmarks: AndroidBench, TerminalBench, and SWE-bench-Verified. Results show that ProcCtrlBench can be instantiated with useful reliability, provides more stable semantics than direct thresholding, and reveals meaningful differences in execution quality that are often overlooked by conventional outcome-based evaluation.

Tool-using agents are increasingly expected to operate across realistic professional workflows, where they must interpret multimodal inputs, coordinate external tools, inspect intermediate artifacts, and revise their actions before producing a final result. Existing benchmarks, however, often evaluate tool use, computer use, and multimodal reasoning in isolation, leaving a gap between benchmark settings and end-to-end omni-modal tool use in the real world. To address this gap, we introduce MM-ToolBench, a benchmark and evaluation harness for task-oriented omni-modal tool use. MM-ToolBench contains 100 executable tasks from two macro task families, Customer Service and Intelligent Creation, covering 20 subcategory slices and supported by 27 MCP servers with 324 tools. The central design of MM-ToolBench is closed-loop multimodal verification: agents must execute tools, inspect rendered or transformed artifacts, and self-correct when outputs fail task-specific requirements. To make such evaluation scalable and verifiable, MM-ToolBench couples MCP-based execution with task-specific grounded evaluators and a semi-automated construction pipeline for scenario discovery, task instantiation, evaluator synthesis, and human audit. Experiments on 15 contemporary agentic models show that MM-ToolBench remains highly challenging: Claude Opus 4.6, commonly regarded as one of the strongest coding-agent models, achieves only 32.0% task success, far below the 94.0% human benchmark. We envision MM-ToolBench as a practical foundation for evaluating and advancing next-generation omni-modal tool-using agents through closed-loop multimodal verification.

Deployed language models are evaluated in a non-stationary environment: model versions, retrieval layers, safety systems, and real-world inputs all change over time. Static bias benchmarks remain useful, but they do not show how models frame newly emerging events for different prompted audiences. We introduce GPF-LIVENEWS, a streaming evaluation protocol and benchmark snapshot for auditing group-conditioned framing in open-ended LLM outputs. The protocol expands fresh BBC/Reuters news anchors across 42 identity labels and seven prompt families, then evaluates response bundles using semantic-sensitivity and sentiment-disparity signals. In a pilot over 12 monitoring runs and 23 hosted models, Policy/Action prompts produce the strongest semantic movement, while sentiment variation is flatter across dimensions and prompt families. The released artifact includes article metadata, prompt templates, instantiated prompts, model-output metadata, score tables, documentation, and reproduction scripts. We interpret all scores as observed-window audit signals for human review, not as permanent fairness rankings or direct proof of harmful bias.

Large language models (LLMs) are increasingly used in the mental health domain, yet it remains unclear how well they capture related biomedical knowledge and how reliably they apply it to clinically salient structured judgments. Here, we present a knowledge-graph (KG)-grounded benchmark for assessing LLMs on mental-health entity recognition, relation judgment, and two-hop reasoning. The benchmark is derived from PrimeKG and comprises nine task families with KG-supported answers and controlled negative options. Experiments across 15 closed- and open-source LLMs reveal a persistent recognition-to-judgment gap: leading models achieve near-ceiling performance on entity typing and on the small relation-typing subset, yet they still struggle with relation prediction and two-hop reasoning. Additionally, short KG-derived snippets benefit some models but degrade performance for others. Moreover, output-format reliability can substantially influence measured performance under constrained multiple-choice settings, highlighting the critical role of response validity in benchmark-based evaluation. MHGraphBench should therefore be interpreted as evaluating agreement with a curated mental-health slice of PrimeKG under a constrained multiple-choice interface, rather than as a direct assessment of real-world clinical safety.

Every AI benchmark operationalizes theoretical assumptions about the capability it claims to assess. When assumptions function as unexamined commitments, benchmarks stabilize the dominant paradigm by narrowing what counts as progress. Over time, narrow evaluation reorganizes capability concepts: architectures and definitions are selected for benchmark legibility until evaluation ceases to track an independent object and instead produces a version of the target defined by its own operational assumptions. The result is a trap: evaluation frameworks treat self-reinforcing assessments as valid, both creating and obscuring structural limits on what the current paradigm can accomplish. We introduce Epistematics, a methodology for deriving evaluation criteria directly from technical capability claims and auditing whether proposed benchmarks can discriminate the claimed capability from proxy behaviors. The contribution is meta-evaluative: an audit procedure, a failure mode taxonomy, and benchmark-design criteria for evaluating capability-evaluation coherence. We demonstrate the procedure through a worked audit of Dupoux et al. (2026), a proposal that revises the dominant paradigm's theoretical assumptions at the architectural level while reproducing them in its evaluation criteria, thereby entrenching the constraint it seeks to overcome in a form the evaluation cannot detect.

LLM-as-a-judge is now the default measurement instrument for open-ended generation, but on the public JudgeBench benchmark even strong instruction-tuned judges barely scrape past random on objective-correctness pairwise items. We introduce RTLC, a three-stage prompting recipe -- Research, Teach-to-Learn, Critique -- that promotes a single black-box LLM into an ensemble-of-thought judge with no fine-tuning, retrieval, or external tools. Stage 1 wraps the input in a fixed pedagogical scaffold porting the Feynman Learning Technique (study $\to$ teach $\to$ find gaps $\to$ simplify) into LLM prompting. Stage 2 draws N=10 independent candidate verdicts at temperature 0.4. Stage 3 acts as its own critic, cross-comparing the candidate set against the original question to emit one critiqued verdict at temperature 0. On JudgeBench-GPT (350 hard pairwise items), Claude 3.7 Sonnet's pairwise accuracy climbs from 64.6% (single-shot vanilla prompt) to 78.6% (RTLC critique-of-10) -- an absolute 14.0-percentage-point gain. RTLC also beats N=10 self-consistency majority voting (77.7%) and a zero-shot first candidate (74.0%). A clean three-step ablation attributes +9.4 pp to the Teach-to-Learn scaffold, +3.7 pp to N=10 marginalisation, and +0.9 pp to explicit critique. We discuss the cost-accuracy frontier (RTLC sits above self-consistency at every working point), the error-budget breakdown across the four JudgeBench categories (knowledge, reasoning, math, coding), and how RTLC composes orthogonally with post-hoc judge-score calibration, with the two interventions compounding multiplicatively in practice.

Kamon (family crests) are an important part of Japanese culture and a natural test case for compositional visual recognition: each crest combines a small number of symbolic choices, but the space of possible descriptions is sparse. We introduce KamonBench, a grammar-based image-to-structure benchmark with 20,000 synthetic composite crests and auxiliary component examples. Each composite crest is paired with a formal kamon description language - "kamon yōgo" - description, a segmented Japanese analysis, an English translation, and a non-linguistic program code. Because each synthetic crest is generated from known factors, namely container, modifier, and motif, KamonBench supports evaluation beyond caption-level accuracy: direct program-code factor metrics, controlled factor-pair recombination splits, counterfactual motif-sensitivity groups under fixed container-modifier contexts, and linear probes of factor accessibility. We include baseline results for a ViT encoder/Transformer decoder and two VGG n-gram decoders, with and without learned positional masks. KamonBench therefore provides a controlled testbed for sparse compositional visual recognition and factor recovery in vision-language models.

We introduce GeoBuildBench, a benchmark designed to evaluate whether large language models and multimodal agents can ground informal natural-language plane geometry problems into executable geometric constructions. Unlike existing geometry benchmarks that focus on answer correctness or static diagram interpretation, GeoBuildBench treats geometry diagram as an interactive construction task: given a textual problem, an agent must generate a domain-specific language (DSL) program to produce a diagram satisfying explicitly specified geometric objects and verifiable constraints. The benchmark features 489 Chinese textbook-style problems, curated through automated filtering and human validation to ensure text-complete, constructible problem specifications. We evaluate several state-of-the-art multimodal models in a bounded iterative setting and show that, despite reasonable success rates, models frequently exhibit structural hallucinations, missing objects, and failures to satisfy geometric constraints, with limited ability to exploit visual and constraint-based feedback for self-correction. These results highlight geometry construction as a rigorous testbed for grounded, executable reasoning beyond textual or visual plausibility. Our benchmark and code are publicly available.

Agent benchmarks have become the de facto measure of frontier AI competence, guiding model selection, investment, and deployment. However, reward hacking, where agents maximize a score without performing the intended task, emerges spontaneously in frontier models without overfitting. We argue that benchmarks must be secure by design. From past incidents of reward hacks, we derive a taxonomy of eight recurring flaw patterns and compile them into the Agent-Eval Checklist for benchmark designers. We condense the insights into BenchJack, an automated red-teaming system that drives coding agents to audit benchmarks and identify possible reward-hacking exploits in a clairvoyant manner. Moreover, we extend BenchJack to an iterative generative-adversarial pipeline that discovers new flaws and patches them iteratively to improve benchmark robustness. We apply BenchJack to 10 popular agent benchmarks spanning software engineering, web navigation, desktop computing, and terminal operations. BenchJack synthesizes reward-hacking exploits that achieve near-perfect scores on most of the benchmarks without solving a single task, surfacing 219 distinct flaws across the eight classes. Moreover, BenchJack's extended pipeline reduces the hackable-task ratio from near 100% to under 10% on four benchmarks without fatal design flaws, fully patching WebArena and OSWorld within three iterations. Our results show that evaluation pipelines have not internalized an adversarial mindset, and that proactive auditing could help close the security gap for the fast-paced benchmarking space.

Recent works have analyzed the impact of individual components of neural networks on gendered predictions, often with a focus on mitigating gender bias. However, mechanistic interpretations of gender tend to (i) focus on a very specific gender-related task, such as gendered pronoun prediction, or (ii) fail to distinguish between the production of factually gendered outputs (the correct assumption of gender given a word that carries gender as a semantic property) and gender biased outputs (based on a stereotype). To address these issues, we curate \gknow, a benchmark to assess gender knowledge and gender bias in language models across different types of gender-related predictions. \gknow allows us to identify and analyze circuits and individual neurons responsible for gendered predictions. We test the impact of neuron ablation on benchmarks for disentangling stereotypical and factual gender (DiFair and the test set of GKnow), as well as StereoSet. Results show that gender bias and factual gender are severely entangled on the level of both circuits and neurons, entailing that ablation is an unreliable debiasing method. Furthermore, we show that benchmarks for evaluating gender bias can hide the decrease in factual gender knowledge that accompanies neuron ablation. We curate GKnow as a contribution to the continuous development of robust gender bias benchmarks.

Optimization modeling translates real decision-making problems into mathematical optimization models and solver-executable implementations. Although language models are increasingly used to generate optimization formulations and solver code, existing benchmarks are almost entirely text-only. This omits many optimization-modeling tasks that arise in operational practice, where requirements are described in text but instance information is conveyed through visual artifacts such as tables, graphs, maps, schedules, and dashboards. We introduce multimodal optimization modeling, a benchmark setting in which models must construct both a mathematical formulation and executable solver code from a text-and-visual problem specification. To evaluate this setting, we develop a solver-grounded framework that generates structured optimization instances, verifies each with an exact solver, and builds both the model-facing inputs and hidden reference files from the same verified source. We instantiate the framework as MM-OptBench, a benchmark of 780 solver-verified instances spanning 6 optimization families, 26 subcategories, and 3 structural difficulty levels. We evaluate 9 multimodal large language models (MLLMs), including 6 frontier general-purpose models and 3 math-specialized models, with aggregate, family-level, difficulty-level, and failure-mode analyses. The results show that the task remains far from solved: the best two models reach 52.1% and 51.3% pass@1, while on average across the six general-purpose MLLMs, pass@1 is 43.4% on easy instances and 15.9% on hard instances. All three math-specialized MLLMs solve 0/780 instances. Failure attribution shows that errors arise both when extracting instance data from text and visuals and when turning extracted data into solver-correct formulations and code. MM-OptBench provides a testbed for solver-grounded, decision-oriented multimodal intelligence.

Authentic school examinations provide a high-validity test bed for evaluating multimodal large language models (MLLMs), yet benchmarks grounded in Japanese K-12 assessments remain scarce. We present a multimodal dataset constructed from Japan's National Assessment of Academic Ability, comprising officially released middle-school items in Science, Mathematics, and Japanese Language. Unlike existing benchmarks based on synthetic or curated data, our dataset preserves real exam layouts, diagrams, and Japanese educational text, together with nationwide aggregated student response distributions (N $\approx$ 900{,}000). These features enable direct comparison between human and model performance under a unified evaluation framework. We benchmark recent multimodal LLMs using exact-match accuracy and character-level F1 for open-ended responses, observing substantial variation across subjects and strong sensitivity to visual reasoning demands. Human evaluation and LLM-as-judge analyses further assess the reliability of automatic scoring. Our dataset establishes a reproducible, human-grounded benchmark for multimodal educational reasoning and supports future research on evaluation, feedback generation, and explainable AI in authentic assessment contexts. Our dataset is available at: https://github.com/KyosukeTakami/gakucho-benchmark

Geoscience intelligence is expected to understand, reason about, and predict earth system changes to support human decision-making in critical domains such as disaster response, climate adaptation and environmental protection. Although current research has shown promising progress on specific geoscience tasks, such as remote sensing interpretation, geographic question-answering, existing benchmarks remain largely task-specific which failing to capture the open-ended real world geoscience problems. As a result, it remains unclear how far current AI systems are from achieving genuine geoscience intelligence. To address this gap, we present \textbf{GeoR-Bench}, a \underline{Bench}mark for evaluating \underline{Geo}science visual \underline{R}easoning through reasoning informed visual editing tasks. GeoR-Bench contains 440 curated samples spanning 6 geoscience categories and 24 task types, covering earth observation imagery and structured scientific representations such as maps and diagrams. We evaluate outputs along three dimensions, including reasoning, consistency, and quality. Benchmark results of 21 closed- and open-source multimodal models reveal that geoscience reasoning remains a critical bottleneck. The highest-performing model achieves 42.7% overall strict accuracy, while the best open-source models only get 10.3%. Notably, the visual consistency and image quality of the outputs frequently surpass their scientific accuracy. Ultimately, these findings indicate that current models generate superficially plausible results but fail to capture underlying earth science processes.

In recent years, a theoretical understanding has rapidly advanced regarding how popular multi-objective evolutionary algorithms (MOEAs) can optimize many-objective problems. However, the benefits of using crossover in many-objective optimization are theoretically not understood, except for specifically designed benchmark functions tuned to particular crossover operators, and still lag significantly behind its practical use. In this paper, we build upon this line of research and present a theoretical runtime analysis of the widely used NSGA-III algorithm on the classical $m$-objective $m$-OneJumpZeroJump function ($m$-OJZJ for short). Our results demonstrate that NSGA-III with crossover optimizes $m$-OJZJ asymptotically faster than NSGA-III without crossover for any number $m$ of objectives for huge parameter regimes. We complement our analysis by providing a lower runtime bound on $4$-OJZJ when crossover is turned off.

Corruption studies, the standard tool for evaluating chain-of-thought (CoT) faithfulness, infer which steps are ``computationally important'' from accuracy loss when steps are corrupted. We show that when benchmark chains end with an explicit terminal answer line, as in GSM8K and MATH, these tests largely measure \emph{answer placement} rather than where intermediate computation is carried out. Using matched GSM8K examples, removing only the final answer statement while preserving all reasoning collapses suffix sensitivity by about $19\times$ for Qwen~2.5-3B ($N{=}300$, $p{=}0.022$). Conflicting-answer prompts, which contain correct reasoning but a wrong explicit final answer, drive accuracy to zero or near-zero at 7B across five open-weight model families; wrong-answer following is strong at 3B--7B and attenuates sharply at larger scales. Replications on MATH, within-stable comparisons at 7B, and suffix-free chains show the same pattern in different guises: corruption sensitivity tracks the location of explicit answer text, not a fixed computational depth in the reasoning. Generation-time probes indicate that final answers are rarely early-determined during generation (${<}5\%$ early commitment), yet consumption-time behavior systematically follows explicit answer text. The confound is therefore largely a readout effect when the chain is consumed. We propose a three-prerequisite protocol (question-only control, format characterization, and an all-position sweep) as a practical minimum for future corruption-based faithfulness studies.

Cell-type-specific marker genes are fundamental to plant biology, yet existing resources primarily rely on curated databases or high-throughput studies without explicitly modeling the supporting evidence found in scientific literature. We introduce PlantMarkerBench, a multi-species benchmark for evaluating literature-grounded plant marker evidence interpretation from full-text biological papers. PlantMarkerBench is constructed using a modular curation pipeline integrating large-scale literature retrieval, hybrid search, species-aware biological grounding, structured evidence extraction, and targeted human review. The benchmark spans four plant species -- Arabidopsis, maize, rice, and tomato -- and contains 5,550 sentence-level evidence instances annotated for marker-evidence validity, evidence type, and support strength. We define two benchmark tasks: determining whether a candidate sentence provides valid marker evidence for a gene-cell-type pair, and classifying the evidence into expression, localization, function, indirect, or negative categories. We benchmark diverse open-weight and closed-source language models across species and prompting strategies. Although frontier models achieve relatively strong performance on direct expression evidence, performance drops substantially on functional, indirect, and weak-support evidence, with evidence-type confusion emerging as a dominant failure mode. Open-weight models additionally exhibit elevated false-positive rates under ambiguous biological contexts. PlantMarkerBench provides a challenging and reproducible evaluation framework for literature-grounded biological evidence attribution and supports future research on trustworthy scientific information extraction and AI-assisted plant biology.

Understanding human mental states from natural behavior is crucial for intelligent systems in the real world. However, most current research focuses on predicting isolated mental state labels, lacking structured annotations of complex interpersonal interactions. To support structured analysis, we introduce MOTOR-Bench, a carefully-designed benchmark with a real-world dataset MOTOR-dataset, containing 1,440 multimodal video clips in collaborative learning scenarios, reflecting key real-world data challenges including natural class imbalance, visual noise, and domain-specific language. Each sample is labeled by educational experts based on self-regulated learning theory. We further evaluate several state-of-the-art multimodal large language models and multi-agent systems in a zero-shot setting on our MOTOR-Bench. However, their performance on this task remains limited, suggesting that existing methods still struggle with structured reasoning from observable behavior to deeper mental states. To address this challenge, we propose a reasoning multi-agent framework, named MOTOR-MAS. It coordinates multiple agents through a structured agent coordination mechanism to infer explicit behaviors, internal cognitions, and psychological emotions. Experimental results show that our MOTOR-MAS outperforms the best single-model benchmark by 15.93 points in Macro-F1 scores for the three labels of behavior, cognition, and emotion, and outperforms the general multi-agent benchmark by 10.2 points in internal cognition prediction.

Large language models (LLMs) have saturated standard medical benchmarks that test factual recall, yet their ability to perform higher-order reasoning, such as synthesizing evidence from multiple sources, remains critically under-explored. To address this gap, we introduce MedMeta, the first benchmark designed to evaluate an LLM's ability to generate conclusions from medical meta-analyses using only the abstracts of cited studies. MedMeta comprises 81 meta-analyses from PubMed (2018--2025) and evaluates models using two distinct workflows: a Retrieval-Augmented Generation (Golden-RAG) setting with ground-truth abstracts, and a Parametric-only approach relying on internal knowledge. Our evaluation framework is validated by a well-structured analysis showing our LLM-as-a-judge protocol strongly aligns with human expert ratings, as evidenced by high Pearson's r correlation (0.81) and Bland-Altman analysis revealing negligible systematic bias, establishing it as a reliable proxy for scalable evaluation. Our findings underscore the critical importance of information grounding: the Golden-RAG workflow consistently and significantly outperforms the Parametric-only approach across models. In contrast, the benefits of domain-specific fine-tuning are marginal and largely neutralized when external material is provided. Furthermore, stress tests show that all models, regardless of architecture, fail to identify and reject negated evidence, highlighting a critical vulnerability in current RAG systems. Notably, even under ideal RAG conditions, current LLMs achieve only slightly above-average performance (~2.7/5.0). MedMeta provides a challenging new benchmark for evidence synthesis and demonstrates that for clinical applications, developing robust RAG systems is a more promising direction than model specialization alone.

Multimodal Large Language Models (MLLMs) may memorize sensitive cross-modal information during pretraining. However, existing MLLM unlearning benchmarks rely on synthetic knowledge injection or complete subject-level deletion, which fail to capture realistic, personalized deletion requests that require fine-grained factual control. In this paper, we introduce PPU-Bench, a real-world and fine-tuning-free benchmark for personalized partial unlearning in MLLMs. PPU-Bench contains 24K multimodal and unimodal samples derived from pre-existing knowledge of 500 public figures under three progressively challenging settings: Complete, Selective, and Personalized unlearning. The benchmark evaluates whether methods can remove target knowledge while preserving non-target facts, model utility, and cross-modal consistency. Extensive experiments show that Complete Unlearning often suppresses visual identity rather than factual knowledge, while Selective and Personalized Unlearning expose significant forget--retain trade-offs and challenges in intra-subject factual boundaries. Robustness analysis under cross-image and prompt-based attacks reveals distinct vulnerabilities across different unlearning settings. Motivated by these findings, we propose Boundary-Aware Optimization (BAO), which explicitly models intra-subject forget-retain boundaries. Experimental results on two representative methods demonstrate that BAO can effectively enforce intra-subject factual boundaries.

While text-to-image models have made strong progress in visual fidelity, faithfully realizing complex visual intents remains challenging because many requirements must be tracked across grounding, generation, and verification. We refer to these requirements as semantic commitments and formalize their lifecycle discontinuity as the Conceptual Rift, where commitments may be locally resolved or checked but fail to remain identifiable as the same operational units throughout the generation lifecycle. To address this, we propose SCOPE, a specification-guided skill orchestration framework that maintains semantic commitments in an evolving structured specification and conditionally invokes retrieval, reasoning, and repair skills around unresolved or violated commitments. To evaluate commitment-level intent realization, we introduce Gen-Arena, a human-annotated benchmark with entity- and constraint-level specifications, together with Entity-Gated Intent Pass Rate (EGIP), a strict entity-first pass criterion. SCOPE substantially outperforms all evaluated baselines on Gen-Arena, achieving 0.60 EGIP, and further achieves strong results on WISE-V (0.907) and MindBench (0.61), demonstrating the effectiveness of persistent commitment tracking for complex image generation.

Text-to-CAD has recently emerged as an important task with the potential to substantially accelerate design workflows. Despite its significance, there has been surprisingly little work on Text-to-CAD evaluation, and assessing CAD model generation performance remains a considerable challenge. In this work, we introduce a new evaluation perspective for Text-to-CAD based on automated testing. We propose CADTestBench, the first test-based benchmark for Text-to-CAD, based on CADTests, executable software tests that verify whether a generated CAD model satisfies the geometric and topological requirements of the input prompt. Using CADTestBench, we conduct comprehensive benchmarking of recent Text-to-CAD methods and further demonstrate that CADTests can also guide CAD model generation, yielding simple baselines that surpass performance of current methods. CADTestBench code and data are available at GitHub and Hugging Face dataset.