{"total":96,"rows":[{"id":"2505.24760","title":"REASONING GYM: Reasoning Environments for Reinforcement Learning with Verifiable Rewards","created":"2025-05-30","cited_by_count":130,"area":null},{"id":"2601.08584","title":"Ministral 3","created":"2026-01-13","cited_by_count":114,"area":null},{"id":"2408.04667","title":"Non-Determinism of \"Deterministic\" LLM Settings","created":"2024-08-06","cited_by_count":92,"area":null},{"id":"2502.06559","title":"Can We Trust AI Benchmarks? An Interdisciplinary Review of Current Issues in AI Evaluation","created":"2025-02-10","cited_by_count":60,"area":null},{"id":"2412.15194","title":"MMLU-CF: A Contamination-free Multi-task Language Understanding Benchmark","created":"2024-12-19","cited_by_count":41,"area":null},{"id":"2502.17521","title":"Recent Advances in Large Langauge Model Benchmarks against Data Contamination: From Static to Dynamic Evaluation","created":"2025-02-23","cited_by_count":38,"area":null},{"id":"2602.16763","title":"When AI Benchmarks Plateau: A Systematic Study of Benchmark Saturation","created":"2026-02-18","cited_by_count":33,"area":null},{"id":"2509.24663","title":"InfLLM-V2: Dense-Sparse Switchable Attention for Seamless Short-to-Long Adaptation","created":"2025-09-29","cited_by_count":32,"area":null},{"id":"2504.15521","title":"The Bitter Lesson Learned from 2,000+ Multilingual Benchmarks","created":"2025-04-22","cited_by_count":31,"area":null},{"id":"2501.08579","title":"LLM-based Human Simulations Have Not Yet Been Reliable","created":"2025-01-15","cited_by_count":30,"area":null},{"id":"2511.04689","title":"Adaptive Testing for LLM Evaluation: A Psychometric Alternative to Static Benchmarks","created":"2025-10-26","cited_by_count":28,"area":null},{"id":"2508.13144","title":"Signal and Noise: A Framework for Reducing Uncertainty in Language Model Evaluation","created":"2025-08-18","cited_by_count":26,"area":null},{"id":"2502.07346","title":"BenchMAX: A Comprehensive Multilingual Evaluation Suite for Large Language Models","created":"2025-02-11","cited_by_count":23,"area":null},{"id":"2601.22966","title":"A Unified View of Attention and Residual Sinks: Outlier-Driven Rescaling is Essential for Transformer Training","created":"2026-01-30","cited_by_count":23,"area":null},{"id":"2410.20245","title":"Improving Model Evaluation using SMART Filtering of Benchmark Datasets","created":"2024-10-26","cited_by_count":21,"area":null},{"id":"2511.09148","title":"LoopTool: Closing the Data-Training Loop for Robust LLM Tool Calls","created":"2025-11-12","cited_by_count":17,"area":null},{"id":"2410.15225","title":"Chasing Random: Instruction Selection Strategies Fail to Generalize","created":"2024-10-19","cited_by_count":13,"area":null},{"id":"2504.07866","title":"Pangu Ultra: Pushing the Limits of Dense Large Language Models on Ascend NPUs","created":"2025-04-10","cited_by_count":13,"area":null},{"id":"2506.13342","title":"Verifying the Verifiers: Unveiling Pitfalls and Potentials in Fact Verifiers","created":"2025-06-16","cited_by_count":13,"area":null},{"id":"2409.18857","title":"Mitigating Selection Bias with Node Pruning and Auxiliary Options","created":"2024-09-27","cited_by_count":12,"area":null},{"id":"2502.14359","title":"Triangulating LLM Progress through Benchmarks, Games, and Cognitive Tests","created":"2025-02-20","cited_by_count":12,"area":null},{"id":"2602.13964","title":"HLE-Verified: A Systematic Verification and Structured Revision of Humanity's Last Exam","created":"2026-02-15","cited_by_count":12,"area":null},{"id":"2601.19532","title":"Benchmarks Saturate When The Model Gets Smarter Than The Judge","created":"2026-01-27","cited_by_count":11,"area":null},{"id":"2411.13757","title":"GenBFA: An Evolutionary Optimization Approach to Bit-Flip Attacks on LLMs","created":"2024-11-21","cited_by_count":10,"area":null},{"id":"2508.09853","title":"STREAM (ChemBio): A Standard for Transparently Reporting Evaluations in AI Model Reports","created":"2025-08-13","cited_by_count":10,"area":null},{"id":"2505.06108","title":"LLMs Outperform Experts on Challenging Biology Benchmarks","created":"2025-05-09","cited_by_count":9,"area":null},{"id":"2503.12440","title":"HKCanto-Eval: A Benchmark for Evaluating Cantonese Language Understanding and Cultural Comprehension in LLMs","created":"2025-03-16","cited_by_count":8,"area":null},{"id":"2505.18502","title":"Knowledge Fusion of Large Language Models Via Modular SkillPacks","created":"2025-05-24","cited_by_count":8,"area":null},{"id":"2508.08237","title":"VGGSounder: Audio-Visual Evaluations for Foundation Models","created":"2025-08-11","cited_by_count":8,"area":null},{"id":"2603.00040","title":"Attn-QAT: 4-Bit Attention With Quantization-Aware Training","created":"2026-02-09","cited_by_count":8,"area":null},{"id":"2505.18102","title":"CapBencher: Give Your LLM Benchmark a Built-in Alarm for Test-Set Overfitting","created":"2025-05-23","cited_by_count":7,"area":null},{"id":"2506.22157","title":"Training Language Model to Critique for Better Refinement","created":"2025-06-27","cited_by_count":7,"area":null},{"id":"2507.23009","title":"Position: Stop Evaluating AI with Human Tests, Develop Principled, AI-specific Tests instead","created":"2025-07-30","cited_by_count":7,"area":null},{"id":"2509.00454","title":"Universal Properties of Activation Sparsity in Modern Large Language Models","created":"2025-08-30","cited_by_count":7,"area":null},{"id":"2605.23912","title":"Raon-Speech Technical Report","created":"2026-04-08","cited_by_count":7,"area":null},{"id":"2510.27543","title":"DialectalArabicMMLU: Benchmarking Dialectal Capabilities in Arabic and Multilingual Language Models","created":"2025-10-31","cited_by_count":6,"area":null},{"id":"2601.21257","title":"MoCo: A One-Stop Shop for Model Collaboration Research","created":"2026-01-29","cited_by_count":6,"area":null},{"id":"2604.13977","title":"How Can We Synthesize High-Quality Pretraining Data? A Systematic Study of Prompt Design, Generator Model, and Source Data","created":"2026-04-15","cited_by_count":6,"area":null},{"id":"2407.12220","title":"Questionable practices in machine learning","created":"2024-07-17","cited_by_count":4,"area":null},{"id":"2511.04234","title":"Reusing Pre-Training Data at Test Time is a Compute Multiplier","created":"2025-11-06","cited_by_count":4,"area":null},{"id":"2511.07464","title":"Motif 2 12.7B technical report","created":"2025-11-07","cited_by_count":4,"area":null},{"id":"2602.05220","title":"Bagpiper: Solving Open-Ended Audio Tasks via Rich Captions","created":"2026-02-05","cited_by_count":4,"area":null},{"id":"2602.10718","title":"SnapMLA: Efficient Long-Context MLA Decoding via Hardware-Aware FP8 Quantized Pipelining","created":"2026-02-11","cited_by_count":4,"area":null},{"id":"2406.11634","title":"The Base-Rate Effect on LLM Benchmark Performance: Disambiguating Test-Taking Strategies from Benchmark Performance","created":"2024-06-17","cited_by_count":3,"area":null},{"id":"2510.14581","title":"Model-agnostic Selective Labeling with Provable Statistical Guarantees","created":"2025-10-16","cited_by_count":3,"area":null},{"id":"2510.15346","title":"When to Ensemble: Identifying Token-Level Points for Stable and Fast LLM Ensembling","created":"2025-10-17","cited_by_count":3,"area":null},{"id":"2606.17574","title":"DeepInsight: A Unified Evaluation Infrastructure Across the Physical AI Stack","created":"2026-06-16","cited_by_count":3,"area":null},{"id":"2607.28109","title":"Beyond Rephrasing: Book-Level Organization Improves Synthetic Textbook Data for Mid-Training","created":"2026-07-30","cited_by_count":3,"area":null},{"id":"2505.24263","title":"Simulating Training Data Leakage in Multiple-Choice Benchmarks for LLM Evaluation","created":"2025-05-30","cited_by_count":2,"area":null},{"id":"2508.08446","title":"OverFill: Two-Stage Models for Efficient Language Model Decoding","created":"2025-08-11","cited_by_count":2,"area":null},{"id":"2602.13680","title":"AllMem: A Memory-centric Recipe for Efficient Long-context Modeling","created":"2026-02-14","cited_by_count":2,"area":null},{"id":"2606.07812","title":"Scaling Participation in Modular AI Systems","created":"2026-06-05","cited_by_count":2,"area":null},{"id":"2608.28113","title":"H-Scale: Hessian-Guided Scale Refinement for NVFP4 Sub-Byte LLM Inference","created":"2026-08-28","cited_by_count":2,"area":null},{"id":"2508.02635","title":"Test Set Quality in Multilingual LLM Evaluation","created":"2025-08-04","cited_by_count":1,"area":null},{"id":"2601.19862","title":"Calibration without Ground Truth","created":"2026-01-27","cited_by_count":1,"area":null},{"id":"2601.22648","title":"UCPO: Uncertainty-Aware Policy Optimization","created":"2026-01-30","cited_by_count":1,"area":null},{"id":"2603.16430","title":"EngGPT2: Sovereign, Efficient and Open Intelligence","created":"2026-03-17","cited_by_count":1,"area":null},{"id":"2605.16893","title":"NGM: A Plug-and-Play Training-Free Memory Module for LLMs","created":"2026-05-16","cited_by_count":1,"area":null},{"id":"2605.18904","title":"Dynamic Model Merging Made Slim","created":"2026-05-17","cited_by_count":1,"area":null},{"id":"2605.21748","title":"RankJudge: A Multi-Turn LLM-as-a-Judge Synthetic Benchmark Generator","created":"2026-05-20","cited_by_count":1,"area":null},{"id":"2605.31455","title":"DRIFT: Decoupled Rollouts and Importance-Weighted Fine-Tuning for Efficient Multi-Turn Optimization","created":"2026-05-29","cited_by_count":1,"area":null},{"id":"2607.02755","title":"Out-of-Distribution Generalization of Risk Aversion in Language Models","created":"2026-07-02","cited_by_count":1,"area":null},{"id":"2607.08779","title":"Signed Symmetric Quantization for Few-Bit Integers","created":"2026-06-12","cited_by_count":1,"area":null},{"id":"2607.18985","title":"Athena-Brain Technical Report: An Efficient Robot Brain for General Intelligence and Embodied Interaction","created":"2026-07-21","cited_by_count":1,"area":null},{"id":"2607.21063","title":"QuantiBias: Benchmarking Quantization-Induced Bias in LLMs","created":"2026-07-23","cited_by_count":1,"area":null},{"id":"2609.02672","title":"oHC: Orthogonal Hyper-Connections on SO(4) via Quaternions","created":"2026-09-02","cited_by_count":1,"area":null},{"id":"2609.06986","title":"Continual Learning Mechanisms Compose for Long-Horizon Memorization","created":"2026-09-07","cited_by_count":1,"area":null},{"id":"2507.06434","title":"Deprecating Benchmarks: Criteria and Framework","created":"2025-07-08","cited_by_count":0,"area":null},{"id":"2602.08294","title":"When Does Context Help? Error Dynamics of Contextual Information in Large Language Models","created":"2026-02-09","cited_by_count":0,"area":null},{"id":"2602.11674","title":"Benchmark Health Index: A Systematic Framework for Benchmarking the Benchmarks of LLMs","created":"2026-02-12","cited_by_count":0,"area":null},{"id":"2604.06829","title":"WRAP++: Web discoveRy Amplified Pretraining","created":"2026-04-08","cited_by_count":0,"area":null},{"id":"2605.01224","title":"Lost in the Tower of Babel: The Adverse Effects of Incidental Multilingualism in LLMs","created":"2026-05-02","cited_by_count":0,"area":null},{"id":"2605.14164","title":"Unsteady Metrics and Benchmarking Cultures of AI Model Builders","created":"2026-05-13","cited_by_count":0,"area":null},{"id":"2605.17273","title":"Position: State-of-the-Art Claims Require State-of-the-Art Evidence","created":"2026-05-17","cited_by_count":0,"area":null},{"id":"2605.30916","title":"Welfare, Improvability, and Variance: A Principal-Agent Approach to Optimal Benchmark Item Aggregation","created":"2026-05-29","cited_by_count":0,"area":null},{"id":"2606.05104","title":"Knowledge Index of Noah's Ark","created":"2026-06-03","cited_by_count":0,"area":null},{"id":"2606.17426","title":"Bounded Difference Concentration for Infinitely Exchangeable Sequences with Applications to AI Benchmark Uncertainty","created":"2026-06-16","cited_by_count":0,"area":null},{"id":"2606.26671","title":"NebulaExp-8B: An Empirical Post-Training Pipeline via Full-Scale Ablation Research","created":"2026-06-25","cited_by_count":0,"area":null},{"id":"2607.02118","title":"Enhancing Fitness Intelligence through Domain-Specific LLM Post-Training","created":"2026-07-02","cited_by_count":0,"area":null},{"id":"2608.15780","title":"Decomposing Staleness in Recommender Systems: A Dual-Filter Framework for Supersession and Decay","created":"2026-08-16","cited_by_count":0,"area":null},{"id":"2609.07102","title":"AstroSpecLM: A Spectrum-Language Model for Evidence-Grounded Astronomical Spectral Analysis","created":"2026-09-07","cited_by_count":0,"area":null},{"id":"2609.15467","title":"Turkish MMLU Pro: Traceable Option Augmentation and Its Validity Limits in Turkish Multiple-Choice Evaluation","created":"2026-09-14","cited_by_count":0,"area":null},{"id":"2609.18284","title":"Made in Hungary: Comments on the performance of generative language models","created":"2026-09-16","cited_by_count":0,"area":null},{"id":"2609.31871","title":"IndustryLLM: Failure-Driven LLM Training for Industrial Procurement","created":"2026-09-25","cited_by_count":0,"area":null},{"id":"2609.33642","title":"Learning to Learn from Context: Synthetic Training from Perturbed Public Documents","created":"2026-09-27","cited_by_count":0,"area":null},{"id":"2609.34754","title":"Draft-KV: Learning Useful Latent Communication Between Language Models","created":"2026-09-28","cited_by_count":0,"area":null},{"id":"2609.36860","title":"IronLLM: Forging Compact Edge-Native Language Models for Real-Time Embodied Intelligence","created":"2026-09-29","cited_by_count":0,"area":null},{"id":"2609.38816","title":"You're Hired: Strategic Model Selection for LLM Collaboration","created":"2026-09-30","cited_by_count":0,"area":null},{"id":"2610.00888","title":"Match the Distribution, Not the Compute: Post-Training Multi-Token Prediction Heads","created":"2026-10-01","cited_by_count":0,"area":null},{"id":"2610.01006","title":"Beyond Answer Confidence: A Controlled Audit of Self-Knowledge in a Black-Box Decision Model","created":"2026-10-01","cited_by_count":0,"area":null},{"id":"2610.10118","title":"YANchor-4B: Effective Long-Horizon Reasoning in O(N) Time with O(1) Memory","created":"2026-10-07","cited_by_count":0,"area":null},{"id":"2610.13630","title":"Multi-Stream Attention Residuals","created":"2026-10-09","cited_by_count":0,"area":null},{"id":"2610.14745","title":"SAGE: Pushing 2-Bit Vector-Quantized LLMs toward Real-World Usability","created":"2026-10-10","cited_by_count":0,"area":null},{"id":"2610.15634","title":"Big-Big-SycoBench: A Benchmark to Detect Sycophancy Through Model Responses to Valid vs. Invalid Arguments","created":"2026-10-11","cited_by_count":0,"area":null},{"id":"2610.19848","title":"OpenTWBench (Open Taiwan Benchmark): Taiwan-Domain Knowledge Evaluation for Large Language Models, and How the Measurement Apparatus Shapes the Score","created":"2026-10-14","cited_by_count":0,"area":null},{"id":"2610.20086","title":"AFCS: Atomic-Fact-Conditioned Synthesis for Factual Knowledge Pretraining","created":"2026-10-14","cited_by_count":0,"area":null}],"next_cursor":null}