Hard

Follow this topic Ask about this topic

Momentum

6 papers in the last four weeks, against 2 the four weeks before. 0.1% of all new papers.

Jul 13Week of Sep 28

Latest papers 58

All topics

What Makes Something Hard(er)? Explaining Question Difficulty in Natural Language
Oct 1, 2026Peng Cui, Qiaoyuan Zheng, Rudolf Debelak +1Hard Item Response Theory
Department of Computer Science ETH Zurich · Machine Learning and Optimization Lab EPFL
Classical Hardness of Learning Functions of Hamiltonians
Oct 1, 2026Sota Hashimoto, Akinori KawachiQuantum Learning Quantum Machine Learning
Graduate School of Engineering, Mie University
HARISSA: Inference-Time Self-Checks for Efficient and Safe Local Language Model Deployment
Sep 29, 2026Kenan Alkiek, Moontae Lee, David Jurgens +1Inference-Time Prefill
School of Information University of Michigan Ann Arbor, MI 48109, USA · SNU-LG AI Research Center Seoul, South Korea
Teaching LLMs to Generate Challenging MILP Instances via Solver Feedback
Sep 29, 2026Jitin Singla, Parikshit Pareek, Pratik Jawanpuria +1Mixed-Integer Programming Hard
IIT Roorkee · IIT Bombay · IIT Delhi
The Uneven Decline of Collective Knowledge Production: Evidence from Stack Overflow After Generative AI
Sep 28, 2026Myokyung Han, Taegyoon Kim, Jinhyuk Yun +1Generative Artificial Intelligence Artificial Intelligence Systems
School of Digital Humanities and Computational Social Sciences, KAIST, 291 Daehak-ro, Yuseong-gu, Daejeon, 34141, Republic of Korea · School of AI Convergence, Soongsil University, 369 Sangdo-ro, Dongjak-gu, Seoul, 06978, Republic of Korea
Not Too Hard, Not Too Easy: Learning from Intermediate States for LLM Structured Reasoning
Sep 27, 2026Hongbo Chen, Guohua Lu, Ting Dang +1LLM Reasoning Strategies Intermediate States
University of Auckland, Auckland, New Zealand · University of Melbourne, Melbourne, Australia
On the Instance Hardness as a Decision Criterion in TinyML Systems
Aug 30, 2026Tobiasz Puslecki, Krzysztof WalkowiakResource-Constrained Edge Devices Computational Cost
Department of Systems and Computer Networks, Wroclaw University of Science and Technology
Long-Horizon AI Research for Grothendieck Constant: A Case Study in Human-AI Mathematical Collaboration
Aug 11, 2026Alan Li, Rahul Saha, Anton Xue +4Artificial Intelligence Research Hilbert Series
Lethe: How Hard Is It to Forget? A Benchmark for Federated Unlearning in Medical Imaging
Aug 2, 2026Shengchao Chen, Ting ShuMedical Imaging Forgetting
University of Technology Sydney · Shenzhen University
CurveShift: Is Agent Progress Scalar? Separating Level from Shape
Jul 31, 2026Hanwen Xing, Pengyun Wang, BingXu Meng +8Agentic Benchmarks Hard
University of Southern California · University of Chicago · University of California, Berkeley +6
Search Hardness-Aware LLM-Based Problem Formulation for Expensive Simulation-Driven Design
Jul 23, 2026Yuchen Li, Handing Wang, Bing Xue +1Automated Heuristic Design Hard
School of Artificial Intelligence, Xidian University, China · School of Engineering and Computer Science, Victoria University of Wellington, New Zealand
Off-Context GRPO: Learning to Reason on Hard Problems using Privileged Information
Jul 21, 2026Priyank Agrawal, Ankur Samanta, Shervin Ghasemlou +4Reinforcement Learning With Verifiable Reward Verifiable Rewards
1Meta AI · 2Columbia University · *Work done at Meta
Entanglement geometry separates circuit cutting, classical hardness, and trainability
Jul 20, 2026Maria Gragera Garces, Sabina Drăgoi, Lirandë PiraQuantum Circuits Variational Quantum Algorithms
Quantum Software Lab University of Edinburgh, UK · ETH ZÜrich, ZÜrich, Switzerland IBM Research, RÜschlikon, Switzerland · Centre for Quantum Technologies, National University of Singapore, Singapore
AI advice suppresses people's willingness to say "I don't know", even when the advice is wrong and accuracy is incentivized
Jul 15, 2026Chiara Marcoccia, Walter Quattrociocchi, Valerio CapraroAdvice Student Misconceptions
Ecole Normale Supérieure, Paris, France · University of Rome “La Sapienza”, Rome, Italy · University of Milan-Bicocca, Milan, Italy
Hierarchical $\mathcal{F}$ -Clustering: Approximation and Hardness of Clustering into Trees and Bounded Diameter Graphs
Jul 14, 2026Michał Szyfelbein, Dariusz DereniowskiClustering Approximation Algorithms
Gda´nsk University of Technology
Are LLMs ready for HardChoices?
Jul 13, 2026Dmitry NikolaevLarge Language Model Bias Ideology
University of Manchester
Semantic Hardness Is Not Visual Hardness: Sign-Aware Hard Negative Mining for Sign Language Retrieval
Jul 10, 2026Junmyeong Lee, Chan Hur, ChangSu Choi +5Sign Language Translation Sign
School of Computing1 · ETRI Medical Informatics Laboratory4 · Graduate School of Culture Technology2 +1
Contravariance Theory: Strong Alignment for Minimal Solutions to Hard Tasks
Jul 9, 2026Dan Yamins, Aran NayebiNeuroevolution Human Brain Networks
Departments of Computer Science and Psychology, and Wu Tsai Neurosciences Institute, Stanford University, Stanford, CA 94305 USA · Machine Learning Department, Neuroscience Institute, and Robotics Institute, Carnegie Mellon University, Pittsburgh, PA 15213 USA
DecompRL: Solving Harder Problems by Learning Modular Code Generation
Jul 2, 2026Juliette Decugis, Fabian Gloeckle, Francis Bach +2Code Generation Hard
FAIR at Meta · Inria · École Normale Supérieure +2
Geometric Signatures of Reasoning: A Spectral Perspective on Task Hardness
Jul 2, 2026Aria Masoomi, Mahsa Bazzaz, Adel Javanmard +1Chain-of-Thought Reasoning Hard
1Northeastern University · University of Southern California · 3Google Research
ShardNet: Training Neural Controllers with Hard, Non-Convex Constraints
Jun 29, 2026Long Kiu Chung, Shreyas KousikSafety Constraints Learning-Based Control
Department of Mechanical Engineering, Georgia Institute of Technology, Atlanta, GA.
HARD-KV: Head-Adaptive Regularization for Decoding-time KV Compression
Jun 27, 2026Yuxuan Yang, Feiyang Ren, Bowen Zeng +4Key-Value Cache Compression LLM Inference Optimization
The State Key Laboratory of Blockchain and Data Security, Zhejiang University
Human vs Machine Mathematical Difficulty on Project Euler: An Experimental Analysis
Jun 20, 2026David Holmes, Johannes SchmittMath Problems Hard
Confusion-Aware Transfer Teacher Curriculum Learning Framework: Disentangling Scoring and Pacing Effects
Jun 16, 2026Savini Kommalage, Sanka Mohottala, Asiri Gawesha +5Adaptive Curriculum Curriculum
Faculty of Computing, Sri Lanka Institute of Information Technology, Sri Lanka · Faculty of Engineering, University of Sri Jayewardenepura, Sri Lanka · Faculty of Engineering, Sri Lanka Institute of Information Technology, Sri Lanka +2
Odds Law: The Decomposition Algebra On How Intelligence Organizes Itself to Solve Difficult Problems Reliably
Jun 14, 2026Hidayet AksuAlgebraic Structures Verification Framework
Harsher on Male? Evaluating LLMs on Gender-Asymmetric Moral Framing Across Diverse Conflict Scenarios
Jun 12, 2026Guangzong Si, Dong Wang, Zhenhao Li +3Moral Reasoning Large Language Model Decisions
University of Science and Technology of China · Eastern Institute of Technology, Ningbo
Sorries Are Not the Hard Part: An Expert-Review Case Study of a Semi-Autonomous Formalization
Jun 11, 2026Vasily Ilin, Brian NugentAutoformalization Formalization
University of Washington, Seattle, WA, USA · University of Utah, Salt Lake City, UT, USA
The Easy, the Hard, and the Learnable: Confidence and Difficulty-Adaptive Policy Optimization for LLM Reasoning
Jun 6, 2026Zhanke Zhou, Xiangyu Lu, Chentao Cao +4Frictive Policy Optimization LLM Reasoning Strategies
1TMLR Group, Department of Computer Science, Hong Kong Baptist University · 2Stanford University · 3Sydney AI Centre, The University of Sydney
Hardness of Multi-Agent Path Finding on Trees: A Unified Approach
Jun 4, 2026Tzvika GeftMulti-Agent Path Finding Np-Hard
Rutgers University, NJ, USA · NJ, USA
Dense Contexts Are Hard Contexts: Lexical Density Limits Effective Context in LLMs
Jun 4, 2026Giovanni Dettori, Matteo Boffa, Danilo Giordano +2Multi-Turn Benchmark Hard
Department of Computer Science Politecnico di Torino Torino, IT 10129 · Department of Computer Science University of Turin Torino, IT 10124
Hedge-Bench: Benchmarking Agents on Hard, Realistic Tasks Pertaining to Financial Reasoning
Jun 2, 2026Eric Cho, Shawn Huang, Alice Lu +1Finance Benchmarks Agentic Benchmarks
1Trata · 2Brigham Young University · 3Osmosis
How Hard Can It Be? Hardness-Aware Multi-Objective Unlearning
Jun 1, 2026Jiangwei Chen, Xinyuan Niu, Rachael Hwee Ling Sim +3Machine Unlearning Catastrophic Forgetting
Department of Computer Science, National University of Singapore, Singapore · Institute for Infocomm Research, Agency for Science, Technology and Research, Singapore · Centre for Frontier AI Research, Agency for Science, Technology and Research, Singapore
The Harder Text Embedding Benchmark (HTEB): Beyond One-dimensional Static Robustness
May 27, 2026Manuel Frank, Haithem AfliHard
Department of Computer Science Munster Technological University
Self-Balancing Gradient Allocation for Heterogeneity-Aware Feature Generation in Click-Through Rate Prediction
May 24, 2026Moyu Zhang, Yun Chen, Yujun Jin +3Balanced Learning Click-Through Rate Prediction
Alibaba Group · Beijing, China
Same Target, Different Basins: Hard vs. Soft Labels for Annotator Distributions
May 20, 2026Mirerfan Gheibi, Gashin GhazizadehLabel Propagation Human Annotators
Independent Researcher.
Semantic Reranking at Inference Time for Hard Examples in Rhetorical Role Labeling
May 18, 2026Anas Belfathi, Nicolas Hernandez, Laura Monceaux +2Automatic Labelling Natural Language Inference
Nantes Université, École Centrale Nantes, CNRS, LS2N, UMR 6004, F-44000 Nantes, France · University of Lorraine, France
From Failure to Feedback: Group Revision Unlocks Hard Cases in Object-Level Grounding
May 15, 2026Yuyuan Liu, Yiping Ji, Anjie Le +6Feedback Hard
Department of Engineering Science, University of Oxford · Australian Institute for Machine Learning, Adelaide University · Technical University of Munich +2
GraphIP-Bench: How Hard Is It to Steal a Graph Neural Network, and Can We Stop It?
May 12, 2026Kaixiang Zhao, Bolin Shen, Yuyang Dai +2Graph Neural Networks Watermarking
University of Notre Dame · Florida State University · University of California, Berkeley
Nice Fold or Hero Call: Learning Budget-Efficient Thinking for Adaptive Reasoning
May 12, 2026Zhaomeng Zhou, Lan Zhang, Junyang Wang +2Reasoning Skills Token Budget Allocation
University of Science and Technology of China · The Chinese University of Hong Kong
Hard to Read, Easy to Jailbreak: How Visual Degradation Bypasses MLLM Safety Alignment
May 8, 2026Zhixue Song, Boyan Han, Yiwei Wang +1Recent Vision-Language Models Multimodal Large Language Models
1AGI Lab, Westlake University, China · University of California, Merced, USA
Verifier-Backed Hard Problem Generation for Mathematical Reasoning
May 7, 2026Yuhang Lai, Jiazhan Feng, Yee Whye Teh +1Mathematical Reasoning LLM Reasoning Strategies
Department of Data Science, City University of Hong Kong · School of Intelligence Science and Technology, Peking University · Work done during Jiazhan’s visiting at the Department of Statistics, University of Oxford. +2
On the Hardness of Junking LLMs
May 6, 2026Marco Rando, Samuel VaiterLarge Language Model Jailbreaks Jailbreak Attacks
Université Côte d’Azur, Inria, CNRS, LJAD, Nice, France · CNRS & Université Côte d’Azur, LJAD, Nice, France
Self-Mined Hardness for Safety Fine-Tuning
May 4, 2026Prakhar Gupta, Garv Shah, Donghua ZhangLarge Language Model Jailbreaks Jailbreaks
University of Michigan
FoR-Net: Focus-on-Regions Network for Semantic Segmentation
May 4, 2026Sheng-Wei Chan, Hsin-Jui Pan, Jen-Shiun ChiangEfficient Semantic Segmentation Semantic Segmentation
Department of Electrical and Computer Engineering, Tamkang University, No.151, Yingzhuan Rd., Tamsui Dist., New Taipei City, 251301, Taiwan
OTSS: Output-Targeted Soft Segmentation for Contextual Decision-Weight Learning
Apr 30, 2026Renjun Hu, Hyun-Soo AhnObjectives Semantic Segmentation
University of Michigan, Ross School of Business
Data-Efficient Indentation Size Effect Correction in Steels Using Machine Learning and Physics-Constrained Neural Network
Apr 30, 2026Radmir Karamov, Tagir KaramovShape-Memory Alloy Hard
Center for Material Technologies · Skolkovo Institute of Science and Technology · Center for Petroleum Science and Engineering
Diffusion Reconstruction towards Generalizable Audio Deepfake Detection
Apr 29, 2026Bo Cheng, Songjun Cao, Xiaoming Zhang +3Environmental Sound Deepfake Detection Audio Deepfake Detection
Department of Electronic and Electrical Engineering, Southern University of Science and Technology, Shenzhen, China · Tencent Youtu Lab, China
Shard the Gradient, Scale the Model: Serverless Federated Aggregation via Gradient Partitioning
Apr 23, 2026Amine BarrakFederated Learning Shard
Department of Computer Science and Engineering, Oakland University, Rochester, MI 48309, USA
HardNet++: Nonlinear Constraint Enforcement in Neural Networks
Apr 21, 2026Andrea Goertzen, Kaveh Alim, Youngjae Min +1Inequality Constraints Structural Constraints
Massachusetts Institute of Technology
QuickScope: Certifying Hard Questions in Dynamic LLM Benchmarks
Apr 20, 2026Taylor Lundy, Narun K. Raman, Kevin Leyton-BrownLarge Language Model Benchmarks Hard
Department of Computer Science University of British Columbia Vancouver, BC
The Harder Path: Last Iterate Convergence for Uncoupled Learning in Zero-Sum Games with Bandit Feedback
Apr 17, 2026Côme Fiegel, Pierre Ménard, Tadashi Kozuno +2Imperfect-Information Games Mirror Descent
ENSAE Paris - CREST, Palaiseau, France · Inria - FairPlay · ENS Lyon, Lyon, France +3
Discover and Prove: An Open-source Agentic Framework for Hard Mode Automated Theorem Proving in Lean 4
Apr 17, 2026Chengwu Liu, Yichun Yin, Ye Yuan +7Theorem Proving Hard
State Key Laboratory for Multimedia Information Processing, School of Computer Science, PKU-Anker LLM Lab, Peking University · Huawei Technologies Co., Ltd. · School of Software & Microelectronics, Peking University +1
Descent Before Hardness: Orbit-Gap Obstructions in Exact Certification
Apr 8, 2026Tristan SimasTractability Certification
McGill University, Montreal, Quebec, Canada
Learning Structural Hardness for Combinatorial Auctions: Instance-Dependent Algorithm Selection via Graph Neural Networks
Feb 16, 2026Sungwoo KangCombinatorial Optimization Algorithm Selection
Department of Electrical and Computer Engineering Korea University Seoul 02841, Republic of Korea
Vision-Based Tactile Sensing for the Perception of the Object's Compliance and Hardness
Oct 14, 2025Muxing Huang, Zibin Chen, Weiliang Xu +5Tactile Deformation
Guangdong Basic Research Center of Excellence for Structure and Fundamental Interactions of Matter, Guangdong Provincial Key Laboratory of Nanophotonic Functional Materials and Devices, School of Optoelectronic Science and Engineering, South China Normal University, Guangzhou 510006, China
Are Targeted Data Poisoning Attacks as Effective as We Think?
Sep 8, 2025William Xu, Chenyu Zhang, Yihan Wang +5Poisoning Hard
Waabi AI · University of Waterloo · Carnegie Mellon University +3
Distributionally Robust Reinforcement Learning with Interactive Data Collection: Fundamental Hardness and Near-Optimal Algorithms
Apr 4, 2024Miao Lu, Han Zhong, Tong Zhang +1Distributional Reinforcement Learning Offline Reinforcement Learning
†Department of Management Science and Engineering, Stanford University. · ‡Center for Data Science, Peking University. · §Department of Computer Science, University of Illinois Urbana-Champaign.
SalamandraTA at WMT 2026 Terminology Shared Task: Hard Examples Are Better Teachers
Date pendingXixian Liao, Maite MeleroTranslation Quality Supervised Finetuning
Barcelona Supercomputer Center

Oct 1, 2026, cs.CLJ/K move · Enter open · S save

What Makes Something Hard(er)? Explaining Question Difficulty in Natural Language

Peng Cui, Qiaoyuan Zheng, Rudolf Debelak, Mrinmaya Sachan

Department of Computer Science ETH Zurich · Machine Learning and Optimization Lab EPFL

Hard Item Response Theory Question Black Box

Open paper AskPDF arXiv

Difficulty is one of the most fundamental properties of a question: it determines whether the question can meaningfully discriminate between models of differing ability. Although a variety of methods can now estimate or predict difficulty automatically, they yield only a single descriptive number, with no account of the underlying factors that make a question difficult in the first place. In this work, we propose a data-driven approach that automatically generates and validates natural-language hypotheses explaining what makes one question harder than another. We first estimate each item's difficulty from the responses of a large pool of LLMs using Item Response Theory. We then sample contrasting sets of easy and hard questions and prompt an LLM to propose candidate explanations of the difference, which are subsequently validated and selected on held-out questions. Experimental results across three datasets spanning mathematical, logical, and commonsense reasoning show that our method produces interpretable and predictive hypotheses. On their own, they predict the difficulty of unseen questions competitively with, or better than, advanced black-box difficulty regressors; used as additional features, they further improve those regressors, implying that they discover difficulty signals that existing models fail to capture. Moreover, we demonstrate that editing questions according to a hypothesis can shift their measured difficulty in the expected direction, indicating that the discovered hypotheses are causally valid difficulty factors rather than post-hoc descriptions. Our approach thus turns a purely descriptive difficulty score into actionable statements.

Momentum

Related topics

Latest papers 58