Incomplete Evidence

A coding agent halfway through an issue has already read much of what a retriever ranks highest. Relevance is scored per passage, but sufficiency belongs to the set: a ranker can fill its budget with variants of one required fact and leave the decision unsupported. We formulate state-conditioned minimal sufficient evidence recovery: given a captured agent state, recover a compact evidence combination that supplies the support its next decision still lacks. SERBench measures this on 500 held-out states from 45 repositories, recording what the agent has seen and crediting only sets that cover every fact the current decision was annotated to require. MSS-Complement treats acquisition as set construction, not ranking. Three semantic calls propose a jointly sufficient set, search for what it lacks, and return 4-8 intact source units within 6,144 tokens. One configuration, fixed on calibration data, recovers a complete set for 73.0% of those states at five items and 80.6% at eight, against 61.4% and 72.4% for Qwen3 embedding with reranking. A matched control ranking by similarity alone reaches 66.6%, placing the gain in the set-level policy, not the computation. From frozen repository source with no gold-derived pool, the lead is 5.0 points. On AMA-Bench it answers from a 76.2% smaller answer prompt, with accuracy 2.08 points above that benchmark's own memory agent. Removing one required group from an otherwise complete set costs 12.3 and 11.1 points of repair-localization precision under two executors. Retrieval for agents is better posed as recovering what a decision lacks than re-ranking what an issue resembles.

Large Language Models (LLMs) are increasingly applied to legal and criminal justice tasks, yet existing work focuses almost exclusively on post-arrest scenarios where the suspect's identity is already known, leaving the critical pre-arrest challenge of inferring suspect characteristics from incomplete evidence largely unexplored. To fill this gap, we introduce the Profiling, Investigation, and Judgment (PIJ), comprising 2,500 real homicide cases from five countries. PIJ evaluates LLMs across three tasks that span the entire criminal investigation pipeline: criminal profiling, which requires abductive reasoning to infer suspect attributes from fragmentary scene evidence, crime process reconstruction, which tests structured information extraction, and sentence prediction, which demands legal deductive reasoning. We evaluate 9 powerful LLMs and find that performance degrades systematically as tasks shift from explicit fact extraction to implicit reasoning over unknown suspect profiles. Categories requiring inferential reasoning, such as motivation and victim-offender relationships, remain the primary bottlenecks. Further analysis reveals substantial gaps between LLMs and human experts, along with pervasive biases in gender, age, and motive attribution. Our findings indicate that pre-arrest inference from incomplete evidence remains an open challenge.

Efficient long-term conversational memory requires retrieving sufficient evidence without indiscriminately expanding the context presented to the language model. This is challenging because relevant evidence may be distributed across multiple sessions, while compression may discard details needed for answering. Different queries therefore require different forms of memory access. To capture these demands, we formulate memory access along two dimensions: discovery breadth, which controls how broadly evidence is searched, and reading fidelity, which controls whether evidence is read in compact form or recovered from the original conversation. Based on this formulation, we introduce JustMem, which stores conversation history as compact atomic memories and adapts memory access along these two dimensions to each query. Specifically, LOOKUP handles local evidence, COMPOSE broadens discovery for distributed evidence, and REPLAY increases reading fidelity for fidelity-sensitive evidence. On LoCoMo and LongMemEval-S, JustMem achieves the highest mean accuracy and retrieval recall among the compared memory systems while using substantially fewer generative-model tokens for memory construction and inference.

Test-time adaptation (TTA) offers many ways to update a deployed model without labels, but choosing the wrong update can make a strong source model worse. Recent methods therefore try to predict which adaptation will work from unlabeled test data. We ask a prior question: does the evidence given to the selector contain enough information to determine the best action at all? We show that this is not guaranteed, even with a perfect selector. If an observation channel makes two deployments look the same while their TTA rankings differ, reliable selection is impossible from that channel; richer evidence can restore the decision only when it resolves the relevant ambiguity. We make this boundary exact in a finite-batch Gaussian TTA model, where doing nothing beats mean recentering for small shifts, recentering wins beyond a unique critical shift, and the boundary shrinks as $1/\sqrt n$. Public benchmark studies on CIFAR-100-C and DomainNet-126 show the same failure mode with modern TTA methods: changing only deployment structure can reverse the oracle action while global order-blind evidence remains unchanged. The result is a practical way to separate two failure modes that are usually mixed together: a weak selector versus an information channel that cannot support the desired decision in the first place.

Agent evaluations face two distinct evidentiary questions: whether a reported claim is recomputable from retained evidence (sufficiency), and whether the retained records cover the committed experiment set (coverage). Generic logs and hash-linked transcripts answer neither reliably. We introduce ClaimReceipt, a claim-relative receipt specification and selective verifier that binds typed transaction evidence to a signed experiment manifest and returns PASS, INVALID, or INCONCLUSIVE per claim. We freeze the specification before implementation (SHA-256 18d109...b81). On 1,392 historical buyer--seller records, a CR-2 verifier reproduces all five manually labeled audit verdicts, exactly replays 600 deterministic and 792 post-generation records, makes every one of 13 declared field groups non-redundant under tested ablations, and returns the expected result on 11/11 semantic faults with 0/8 false positives. We then run a separate prospective CR-3 epoch: 30 assignments are committed before inference, terminal receipts are signed and chained, and private evidence is encrypted for an auditor. Complete evidence yields coverage and accounting PASS; withholding one terminal receipt returns INCONCLUSIVE_COVERAGE, while withholding all private openings preserves coverage and protocol verification but makes economic claims inconclusive, exactly matching a preregistered prediction. Receipt instrumentation adds 0.021% of model-inference time and 9.9 KB per transaction. A specification-legibility probe indicates that our own frozen specification is not yet unambiguous to an independent reader. Claim verification therefore requires both claim-sufficient evidence and a committed universe against which omissions become visible.

A failed retrieval-augmented generation (RAG) answer can be consistent with several unseen responses to evidence repair. We introduce Pair-ID, an offline audit that holds one query, retrieval state, and reader constant, then crosses two operations, adding missing support and deleting verified nonsupport, to measure a same-failure counterfactual response vector. A complete funnel over 19,981 benchmark queries identifies 11,105 eligible Qwen failures, from which a prospectively fixed SHA-256 ordering selects 1,200 before generating any sampled response. Among 1,190 regenerated-valid failures, support addition repairs 197/600 JOINT cases (0.328, 95% CI [0.292, 0.367]), and deletion repairs 162/1,190 cases (0.136, 95% CI [0.117, 0.155]); length- and position-matched shams retain semantic contrasts of 0.223 and 0.101. The original view carries partial predictive signal for individual response cells (macro AUROC 0.678; Brier 0.152 versus 0.160 for a marginal baseline), but exact-vector accuracy, 0.637, does not exceed the 0.646 majority-vector baseline, and vector macro-F1 is 0.170. Across four readers, both marginal sensitivities recur, while pooled exact-vector agreement is 0.675-0.765 and JOINT-only agreement falls to 0.538-0.691. These results show that evidence sensitivity occurs at meaningful rates in the hash-selected eligible-failure sample, is only partially predictable from the observed failure, and is conditional on the reader. The evidence supports a frame-scoped offline response audit, not an information-theoretic impossibility result, reader-independent taxonomy, or runtime repair policy.

Emergency triage requires reliable decisions within a short time period. However, the available electronic health record (EHR) data, including structured data and clinical text, are often incomplete, unreliable, and inconsistent. This makes machine learning (ML)-based triage prediction more challenging, as existing ML models typically rely on complete and reliable EHR data to accurately predict patients' acuity levels. To address this, we propose confidence- and reliability-aware selective triage (CRS-Triage) to predict patients' acuity levels with a confidence score. By comparing the confidence score with a predefined threshold, CRS-Triage can selectively determine whether the model should make the decision or defer the case. Specifically, CRS-Triage separately evaluates the reliability of structured data and clinical text and then jointly considers the consistency between the two modalities to estimate the confidence of each prediction. Moreover, to reduce the risk of missing high-acuity patients, namely under-triage, CRS-Triage prefers to assign patients slightly higher acuity levels, namely over-triage, by penalizing under-triage errors. Experiments on the MIMIC-IV-ED dataset show that CRS-Triage achieves strong predictive performance. It also provides a better risk-coverage trade-off and remains reliable when the available EHR data are incomplete, degraded, or inconsistent across modalities.

Medical LLMs are often evaluated by whether they select the correct diagnosis, but diagnostic accuracy alone does not show whether the model used the case evidence appropriately. We present a behavioral audit of evidence use in medical diagnosis. For each case, we decompose patient information into evidence units, score candidate diagnoses under controlled evidence subsets, and mine low-order interactions in diagnostic margins. Because medical evidence is diagnosis-relative, the audit separates interaction discovery from failure assignment: large or negative interactions can reflect plausible differential diagnosis, while suspicious interactions require robustness checks and clinical review. We evaluate five open-weight LLMs on DDXPlus, CupCase, and MedCase. Across datasets, faithful support and differential conflict or cancellation account for most interaction strength, showing that many evidence interactions are clinically plausible rather than failures. In a DDXPlus-focused blinded five-reviewer 130-item enriched review sample, invalid or shortcut-like cases concentrate in negated or absent findings and clinically local evidence. These results show that accuracy can hide candidate evidence-use failures and motivate role-aware audits for medical LLM evaluation.

Independent agents that reason in latent space can share computed state as key-value cache fragments rather than text. Merged by a conflict-free replicated data type, these fragments form a store that converges under any delivery order or duplication. Yet a later query, unknown at encode time, cannot reliably read the merged cache: colocated fragments interfere, so colocation is not addressability. MaSRead addresses the read to content. It routes through opaque keyed tag sets derived from fragment words and decodes each selected fragment under a hard attention mask that hides the rest. Under lexical connectivity, a graph walk reaches the fragments required by a multi-hop query. Across chain, pipeline, symmetric, hub, and natural-language stores, MaSRead recovers visited fragments in isolation, remains effective as unrelated fragments accumulate, and transfers to another model family. After routing, materialized decoding depends on fragment length rather than total store size; end-to-end work still includes store-dependent routing and one read per visited fragment. The limits are explicit: lexical routing can miss disconnected evidence, and answer composition remains bounded by the frozen reader. Thus a replicated latent store becomes selectively readable for later queries when the needed fragments connect to the query through content.

Self-supervised learning (SSL) models are widely used as feature extractors for state-of-the-art audio deepfake detection, but it remains unclear how to directly and quantitatively connect what SSL models capture to detection decisions. To address this gap, we propose Evidence Subspace Projection, a method that represents both evidence factors (e.g., attack category, codec, gender, transmission) and authenticity labels in a shared space constructed from SSL models' neuron activation patterns. By projecting the decision vector onto each evidence subspace, we obtain a scalar ratio that quantifies the explanatory power of each evidence type. We evaluate SSL models in raw, fine-tuned, and post-trained settings on multiple datasets. The results confirm findings from established studies, validating the proposed method, and reveal new insights into model behavior.

Unified visual anomaly detection seeks to train a single detector that can be deployed across categories, domains, and application scenarios. In the few-shot transfer regime, the key challenge is to estimate an episode-specific boundary for an unseen target category from a small support set. Existing approaches mainly infer this boundary from normal-side evidence and provide limited abnormal-side evidence for deployment-specific tolerance. Within the normal side, they often struggle to jointly capture local correspondences and global support-query relations, making their boundaries less reliable for unseen anomalies. To address these issues, we propose UniVAD v2, a two-sided support-conditioned boundary construction framework for unified visual anomaly detection. Built on the component-patch divide-and-conquer framework of UniVAD, UniVAD v2 strengthens the normal side with an Optimal Transport-based Relational Modeling module (OTRM), which complements retrieval with support-query matching through transport-style allocation, and an Adaptive Coordination mechanism for Retrieval and Relational Modeling (ACRRM), which estimates episode-conditioned reliabilities to fuse the two sources of evidence. On the abnormal side, a Few-Shot Abnormal Reference module (FAR) converts optional abnormal references into rejection-side evidence for boundary adjustment. Experiments on six datasets spanning industrial, logical, and medical anomaly detection demonstrate strong cross-domain generalization. Under the 1N-shot protocol, UniVAD v2 improves the mean image-level AUC over UniVAD from 83.0% to 84.5%, and further reaches 85.7% in the 1N+1A-shot setting. On the MVTec-AD Severity Split (MVTec-AD-SS), UniVAD v2 achieves 96.2% image-level AUC and 96.9% pixel-level AUC, showing that abnormal references enable controllable boundary customization without retraining.

Collaborative perception serves as a pivotal solution to enhance the perception capability of individual agents in autonomous driving, where a core challenge lies in seeking reliable evidence to quantify and weight the contribution of each participating agent. Existing methods typically rely on a confidence map, which is co-trained with the detection head, but it is inherently correlated with the detection results and thus fails to provide unbiased physical evidence. Furthermore, how to deeply integrate evidence into the cooperative fusion process remains an open question. To address these issues, this paper first proposes an uncertainty map, a physically grounded and unambiguous metric for evaluating perception quality. This map is directly supervised by real-time sensor signals, i.e., LiDAR point density, ensuring decoupling from detection noise and thereby providing physical scenario-aware evidence for weighting agent contribution. Based on this map, we develop the Uncertainty-Enhanced Collaborative Perception (UECP) framework, centered on the Uncertainty-Aware Pyramid Fusion (UAPF) module. UAPF uses a coarse-to-fine strategy, with two key components: Uncertainty-Weighted Downsampling (UWD) for high-fidelity feature preservation, and Uncertainty-Guided Residual Fusion (UGRF) to reinforce ego features, suppressing noise and ensuring robust fusion. Extensive experiments on real-world datasets show UECP outperforms state-of-the-art methods in effectiveness and robustness by embedding the uncertainty map into fusion. Code will be publicly available.

Language models can generate plausible rationales for their predictions, but these explanations may not faithfully represent the model's internal reasoning. We propose verifier-coupled reasoning, a framework that inserts inline claims into reasoning traces and trains an auxiliary consistency head to predict programmatic verifier outputs from rationale-span hidden states. The central finding is a gap between decodability and faithfulness: consistency training reliably makes verifier information decodable from rationale representations, but decodability does not guarantee faithful generation. In LeanCheck (formal theorem proving), rationale-only and proof-only pooling achieve perfect directional separation under counterfactual conflict. In KataGo (Go engine), commentary spans encode 10-way win-rate buckets at 81% accuracy. Yet in a code setting, the model achieves 98.6% coupling while its generated explanations remain unfaithful: fluent prose with correct structured claims, but describing unrelated algorithms; a controlled pretrained-vs-from-scratch comparison shows the gap is not capacity-driven. Synthetic activation patching confirms causal influence (73-89% vs. 31% baseline), FEVER reveals that evidence-only pooling isolates genuine evidence sensitivity at the cost of raw accuracy, and per-claim analysis shows that consistency loss disproportionately benefits fine-grained claims over binary ones. These results establish that consistency losses are effective diagnostics and representation-shaping tools, but not sufficient conditions for faithful reasoning.

Audit risk assessment increasingly benefits from combining heterogeneous evidence sources, yet existing approaches typically produce point predictions without quantifying how well different evidence streams agree. We propose UMAR (Uncertainty-Aware Multi-Agent Risk Assessment), a framework that employs three specialized agents: an MD&A Text Agent, a Financial Ratio Agent, and a CAM Agent, each producing independent risk scores with calibrated uncertainty estimates. An Uncertainty Aggregator based on Dempster-Shafer evidence theory fuses these scores while explicitly measuring inter-agent conflict. We evaluate UMAR on a U.S. dataset of 3,200 firm-year observations from SEC 10-K filings (2019-2023), with financial restatement as the target label. Experimental results show that UMAR achieves an AUROC of 0.782 and a PR-AUC of 0.341, outperforming logistic regression, XGBoost, FinBERT, and single-agent and dual-agent LLM baselines. UMAR attains the lowest expected calibration error (ECE = 0.052) among all methods and identifies evidence-conflict patterns that correlate with actual restatement risk, offering auditors potentially actionable and interpretable risk signals.

Spatial reasoning is fundamental to robotics, autonomy, and embodied AI, yet modern vision-language models (VLMs) remain unreliable on metric distance queries. A common assumption is that consistent predictions across viewpoints reflect geometric grounding. We test this assumption and find the opposite: leading VLMs often produce view-invariant and consistent answers even when those answers are incorrect, indicating weak coupling between predictions and viewpoint-specific visual evidence. We introduce \textbf{ViewDiag}, a controlled multi-view evaluation protocol built from Hypersim, ScanNet, and KITTI360, comprising 176 object-pair tracks across 80 scenes with 2--10 views per track. The protocol evaluates models along three axes: metric accuracy, distributional concentration, and internal collapse, the last of which is assessed using a latent feature probe. Across diverse models, we observe a consistent pattern of high prediction stability paired with substantial error, clustering in a regime characterized by strong consistency but low accuracy. \noindent These results challenge the common use of cross-view consistency as a proxy for geometric understanding. Instead, we show that stable predictions may reflect prior-driven collapse rather than evidence-sensitive reasoning. ViewDiag provides a controlled benchmark and diagnostic framework for evaluating whether spatial VLMs are not only accurate, but also meaningfully coupled to visual evidence.

Emerging personal AI agents are moving toward persistent, multi-source memory. This creates an evaluation problem: systems must decide how to use conflicting or incomplete evidence; they cannot just retrieve facts from one clean history. Existing benchmarks rarely show whether an error came from the evidence given to a method or from the method's conflict-resolution step. We study this as selective QA over conflicting multi-source personal memory: systems answer based on conflicting, sometimes incomplete sources, or abstain when evidence is insufficient. We develop a benchmark containing 18 question templates across 8 reasoning types, 480 personas, 4 random seeds, and 34,560 instances, with controlled source distortions and deterministic ground truth. We evaluate the performance of baselines without access to any source, access to a single source, structured fusion methods, and frontier LLMs. The best trained fusion resolver reaches 80.3% accuracy, while the strongest prompt-only LLM baseline reaches 70.0%. With abstention, the same resolver reaches 85.3% selective accuracy at 78.3% coverage and the best LLM reaches 71.0% selective accuracy at 95.4% coverage. Different models have different strengths across reasoning types. We release the data, code, cached model outputs, and data-generating process for reuse.

We highlight a failure mode of large reasoning models on questions with insufficient information: models may recognize that a problem is under-specified, yet still continue reasoning and produce unsupported final answers instead of abstaining. We formalize this mismatch as the detection-to-abstention gap, where detected insufficiency fails to translate into final abstention. This gap is especially concerning in high-risk domains such as medical AI, where answers based on incomplete evidence can be more harmful than refusal. To close this gap, we propose Judge-Then-Solve (JTS), a trajectory-level reasoning-control framework that trains models to make an explicit answerability commitment before solution generation. Rather than treating abstention as a final-answer style, JTS casts it as a control decision: the model either proceeds to solve or terminates early based on its answerability judgment. We instantiate this policy through supervised warm-up and missing-premise reinforcement learning with consistency and length-shaping rewards. Experiments on dense and MoE reasoning models show that JTS substantially improves reliable abstention across datasets and pushes Abstention@Detection (A@D) to near-saturation, indicating that models not only detect missing information but also act on that detection. By terminating unanswerable trajectories immediately after the answerability judgment, JTS reduces unnecessary reasoning and improves inference efficiency when continued deliberation would amplify unsupported assumptions. We also observe that missing-premise training can alter reasoning behavior on difficult but answerable problems, reducing unproductive self-reflection. These results suggest that abstention under insufficient information is a key form of reasoning control for deploying reasoning models safely and efficiently.

Multilingual fact verification requires evidence that is both relevant and sufficiently complete for reliable factuality prediction. However, existing systems often rely on search snippets, sentence-level evidence, or locally segmented passages, which can miss decisive context and produce fragmented evidence. To overcome these limitations, we propose SEEK, a Semantic Evidence Extraction with an adaptive chunKing framework that constructs coherent evidence chunks from full fact-checking articles by identifying semantic topic transitions and preserving local verification context. The constructed chunks are encoded using a multilingual encoder and then multilingual LLMs are finetuned using LoRA adapter for veracity prediction. Experiments on X-FACT and RU22Fact show that SEEK improves macro-f1 by up to 10% over semantic chunking, 19% over sentence chunking, and 20% over search-snippet baselines. Evidence completeness and significance analyses further show that SEEK preserves richer verification context and enables more reliable multilingual fact-checking.

Evidence absence is not evidence insufficiency, but fact verification benchmarks can make them observationally similar. The Not Enough Information (NEI) label is often operationalized through constructed evidence conditions, and that choice silently determines what a verifier learns. We introduce NEI-CAP, a construction-aware diagnostic protocol for insufficient-evidence evaluation. Each NEI example carries the construction family that produced it; NEI-CAP audits shortcut cues, validates hard cases through human adjudication, and tests whether competence transfers across constructions. We instantiate the protocol on SciFact, with FEVER and HoVer as bounded external controls. Across these settings, NEI competence does not transfer reliably: encoder verifiers and an instruction-tuned decoder trained on shortcut-prone constructions fail to recognize semantically related insufficient evidence, and mixed-construction training narrows but does not close the gap. Fixed-claim diagnostics further show that the evidence condition shifts confidence in the reference Support/Refute label, not only NEI recall, so an aggregate NEI score can hide which problem a model has actually solved. We therefore recommend reporting the construction family alongside the score, and distill the results into a checklist for benchmarks that carry an insufficient-evidence label.

Evidence construction--the stage that determines which passages reach the language model before generation begins--is evaluated paradigm by paradigm, leaving practitioners with no principled way to diagnose which organization strategy fails, where, or why. We introduce AuthTrace, a diagnostic benchmark built on thematically dense single-author corpora where near-miss distractors share style, topic, and vocabulary with the required evidence. AuthTrace provides explicit quoted evidence, exact fan-in annotation, and a unified pack-level protocol measuring evidence recall, evidence precision, and answer correctness. A fan-in gradient--the number of source documents required to support the answer--serves as the primary diagnostic axis, enabling controlled comparison across retrieval, memory, graph, and structured-evidence paradigms. Evaluating eight systems across two QA models, we find that evidence recall is the strongest observed predictor of answer correctness under the primary reader-judge pair (r = 0.96); most failures stem from missing evidence rather than answer synthesis. Fan-in further exposes paradigm-specific collapse patterns: flat retrieval degrades 2-3x faster than thematically organized evidence construction. These results show fan-in decomposition to be a reusable diagnostic lens for identifying where evidence-construction systems fail and which paradigm best serves a given workload.

As demand for mental health care outpaces clinician-delivered assessment, scalable screening tools are increasingly needed. Large language models (LLMs) may identify psychiatric risk from patient narratives, but their reliability across diagnoses, demographic subgroups, and evidence-use patterns remains uncertain. We introduce a SCID-anchored benchmark of 555 semi-structured experiential interviews paired with diagnostic reference labels for anxiety disorder, major depressive disorder, post-traumatic stress disorder, and any current mental health disorder. Using zero-shot task-specific prompting, we evaluated five state-of-the-art LLMs and examined whether false-negative errors reflected missed psychiatric evidence or differential weighting of symptom, functional-impairment, and protective-context cues. Performance varied across tasks and models, with accuracy ranging from 0.49 to 0.86 and Matthews correlation coefficients from 0.16 to 0.38. GPT-4.1 Mini and GPT-5 Mini showed the most consistent disorder-specific accuracy. Subgroup analyses found higher depression-classification accuracy among male than female participants, no consistent age-related pattern, and modest non-uniform variation across race strata. Evidence-integration analyses showed that false-negative anxiety and PTSD classifications often contained explicit symptom evidence but were accompanied by preserved functioning, coping ability, or social support. Functional-impairment evidence shifted model outputs toward positive classifications, whereas protective-context evidence shifted outputs away. These findings suggest that LLMs may support scalable psychiatric screening, but their tendency to discount symptom evidence in the presence of preserved functioning or protective context requires careful validation before clinical deployment.

Biomedical retrieval-augmented large language models (LLMs) often face evidence that is incomplete, misleading, or internally contradictory, yet evaluation usually emphasizes answer accuracy under helpful context rather than reliability under conflict. Using HealthContradict, we evaluate six open-weight LLMs under five controlled evidence conditions: no retrieved context, correct-only context, incorrect-only context, and two mixed conditions containing both correct and contradictory documents in opposite orders. In this conflicting-evidence order contrast, where the same two documents are both present and only their order is reversed, accuracy drops for every model and 11.4%--25.2% of predictions flip. To support abstention in these difficult cases, we also evaluate a conflict-aware abstention score that combines model confidence with a detector of evidence conflict. In the two hardest conditions, this score improves selective accuracy over confidence-only, with mean gains of 7.2--33.4 points in incorrect-only (IC') and 3.6--14.4 points in incorrect-first conflicting (ICC') conditions across 75%, 50%, and 25% coverage. These results show that conflicting biomedical evidence is both an uncertainty and robustness problem and motivate evaluation and abstention methods that explicitly account for evidence disagreement.

Marketplace platforms routinely evaluate pricing and allocation policies using logged observational data, yet strong offline performance does not imply that a policy is safe to deploy. In real-time bidding (RTB) marketplaces, reserve-price and floor-policy changes affect not only revenue but also fill, advertiser value, budget pacing, and competition across auctions, creating feedback and interference. The central problem is therefore not to estimate whether a policy improves an offline metric, but to determine whether the available evidence justifies direct launch or only further validation. In this regard, we propose a support-aware decision-support system (DSS) that distinguishes promising from actionable evidence. The framework integrates replay, support-aware off-policy evaluation (OPE), conservative lower-bound ranking, multi-sided guardrails, out-of-time validation, sensitivity analysis, and interference-aware validation design into a claim-preserving pipeline that outputs a launch-readiness classification rather than a single performance estimate. Applying the framework to iPinYou-style RTB logs, we identify a margin-gated floor policy as the leading candidate, with a 47.7% replay yield lift, a 45.8% conservative lower-tail lift, and stable out-of-time performance. However, the framework does not recommend direct launch. A decision-rule ablation shows that simplified pipelines select the same policy but incorrectly recommend deployment, leaving key causal assumptions unresolved. In contrast, the proposed DSS selects the same policy but changes the action to online validation, reflecting missing evidence on propensities, bidder response, and interference. Overall, the contribution is a reproducible DSS protocol that prevents decision overclaim under partial identification and converts offline evaluation into an auditable, action-oriented recommendation.

Enterprise agents increasingly operate inside scoped retrieval systems, delegated workflows, and policy-constrained evidence environments. In these settings, access control can be enforced correctly while the system still produces an answer that appears complete even though material evidence lies outside the caller's authorization boundary. This paper introduces Partial Evidence Bench, a deterministic benchmark for measuring that failure mode. The benchmark ships three scenario families -- due diligence, compliance audit, and security incident response -- with 72 tasks total, ACL-partitioned corpora, oracle complete answers, oracle authorized-view answers, oracle completeness judgments, and structured gap-report oracles. It evaluates systems along four surfaces: answer correctness, completeness awareness, gap-report quality, and unsafe completeness behavior. Checked-in baselines show that silent filtering is catastrophically unsafe across all shipped families, while explicit fail-and-report behavior eliminates unsafe completeness without collapsing the task into trivial abstention. Preliminary real-model runs show model-dependent and scenario-sensitive differences in whether systems overclaim completeness, conservatively underclaim, or report incompleteness in an enterprise-usable form. The benchmark's broader contribution is to make a governance-critical agent failure measurable without human judges or contamination-prone static corpora.

Compositional inference - the decomposition of observations into an unknown number of latent components - is central to perception and scientific data analysis. Attention-based models perform well when components are approximately separable, as in object-centric vision. Under additive superposition, however - where multiple components contribute to every observation - we identify a structural failure mode we term slot collapse: multiple slots converge to the same dominant component while weaker ones remain unrepresented. We trace this to a general limitation: attention is memoryless with respect to explained evidence. All slots repeatedly operate on the same input without accounting for what has already been explained, so gradients are dominated by the strongest component, inducing shared fixed points across slots. As a result, attention fails to enforce non-redundant allocation under additive superposition. We address this by introducing residual evidence modeling, instantiated via evidence depletion - a minimal modification combining multiplicative depletion with an attention bias. Controlled ablations show that parallel attention, sequential processing alone, and loss-based regularization fail to resolve collapse; evidence depletion, which adds residual state to sequential attention, consistently succeeds. Across synthetic benchmarks and real-world audio mixtures (FUSS), evidence depletion reduces slot collapse by up to an order of magnitude, generalizing beyond synthetic settings. On gravitational-wave source inference for the ESA/NASA LISA mission, under identical architectures, data, and losses, standard attention fails while evidence depletion prevents collapse and enables multi-source posterior estimation. These results show that under additive superposition, residual evidence tracking is the operative ingredient for preventing collapse and enabling compositional inference.

A well-known limitation of AI systems is presumptuousness: the tendency of AI systems to provide confident answers when information may be lacking. This challenge is particularly acute in legal applications, where a core task for attorneys, judges, and administrators is to determine whether evidence is sufficient to reach a conclusion. We study this problem in the important setting of unemployment insurance adjudication, which has seen rapid integration of AI systems and where the question of additional fact-finding poses the most significant bottleneck for a system that affects millions of applicants annually. First, through a collaboration with the Colorado Department of Labor and Employment, we secure rare access to official training materials and guidance to design a novel benchmark that systematically varies in information completeness. Second, we evaluate four leading AI platforms and show that standard RAG-based approaches achieve an average of only 15% accuracy when information is insufficient. Third, advanced prompting methods improve accuracy on inconclusive cases but over-correct, withholding decisions even on clear cases. Fourth, we introduce a structured framework requiring explicit identification of missing information before any determination (SPEC, Structured Prompting for Evidence Checklists). SPEC achieves 89% overall accuracy, while appropriately deferring when evidence is insufficient -- demonstrating that presumptuousness in legal AI is systematic but addressable, and that doing so is a necessary step towards systems that reliably support, rather than supplant, human judgment wherever decisions must await sufficient evidence.

When does an LLM controller outperform rule-based traversal for knowledge graph exploration? We study this question through RLM-on-KG, a retrieval system that treats an LLM as an autonomous navigator over an RDF-encoded mention graph for grounded question answering. Unlike GraphRAG pipelines that rely on offline LLM indexing, RLM-on-KG performs entity-first, multi-hop exploration at query time using deterministic graph construction and a fixed tool set. Our central finding is a conditional advantage: the value of LLM control depends on evidence scatter and tool-calling sophistication. The paper's core claim is LLM control versus heuristic traversal, not a generic win over GraphRAG. On GraphRAG-Bench Novel (519 questions), Gemini 2.0 Flash achieves +2.47 pp F1 over a rule-based heuristic baseline (p < 0.0001), but only +0.16 pp over a GraphRAG-local variant (not significant). With a stronger controller, Claude Haiku 4.5, the gain over heuristic grows to +4.37 pp (p < 0.001) and extends to a +2.42 pp significant improvement over GraphRAG-local (p < 0.001). The gain is largest when gold evidence is scattered across 6-10 chunks (+3.21 pp) and smallest for concentrated evidence (+1.85 pp). Cross-scale validation on MuSiQue confirms that the LLM-over-heuristic advantage transfers, with expected attenuation on smaller per-question graphs. The core architectural insight is the separation of candidate discovery from ranking: the LLM adds value through exploration breadth, while final evidence selection is best handled by pure vector re-ranking. Beyond retrieval, exploration traces provide a proposed stress-test harness for structured data quality, yielding diagnostics for coverage, connectivity, provenance, and queryability.

Effective abstention (EA), recognizing evidence insufficiency and refraining from answering, is critical for reliable multimodal systems. Yet existing evaluation paradigms for vision-language models (VLMs) and multi-agent systems (MAS) assume answerability, pushing models to always respond. Abstention has been studied in text-only settings but remains underexplored multimodally; current benchmarks either ignore unanswerability or rely on coarse methods that miss realistic failure modes. We introduce MM-AQA, a benchmark that constructs unanswerable instances from answerable ones via transformations along two axes: visual modality dependency and evidence sufficiency. Evaluating three frontier VLMs spanning closed and open-source models and two MAS architectures across 2079 samples, we find: (1) under standard prompting, VLMs rarely abstain; even simple confidence baselines outperform this setup, (2) MAS improves abstention but introduces an accuracy-abstention trade-off, (3) sequential designs match or exceed iterative variants, suggesting the bottleneck is miscalibration rather than reasoning depth, and (4) models abstain when image or text evidence is absent, but attempt reconciliation with degraded or contradictory evidence. Effective multimodal abstention requires abstention-aware training rather than better prompting or more agents.