MathNet delivers the largest multilingual Olympiad math dataset and benchmarks where models like Gemini-3.1-Pro reach 78% on solving but embedding models struggle on equivalent problem retrieval, with retrieval augmentation yielding up to 12% gains.
hub Mixed citations
Omni-MATH: A Universal Olympiad Level Mathematic Benchmark For Large Language Models
Mixed citation behavior. Most common role is background (62%).
35
Pith papers citing it
Background
62% of classified citations
abstract
Recent advancements in large language models (LLMs) have led to significant breakthroughs in mathematical reasoning capabilities. However, existing benchmarks like GSM8K or MATH are now being solved with high accuracy (e.g., OpenAI o1 achieves 94.8\% on MATH dataset), indicating their inadequacy for truly challenging these models. To bridge this gap, we propose a comprehensive and challenging benchmark specifically designed to assess LLMs' mathematical reasoning at the Olympiad level. Unlike existing Olympiad-related benchmarks, our dataset focuses exclusively on mathematics and comprises a vast collection of 4428 competition-level problems with rigorous human annotation. These problems are meticulously categorized into over 33 sub-domains and span more than 10 distinct difficulty levels, enabling a holistic assessment of model performance in Olympiad-mathematical reasoning. Furthermore, we conducted an in-depth analysis based on this benchmark. Our experimental results show that even the most advanced models, OpenAI o1-mini and OpenAI o1-preview, struggle with highly challenging Olympiad-level problems, with 60.54\% and 52.55\% accuracy, highlighting significant challenges in Olympiad-level mathematical reasoning.
hub tools
citation-role summary
background 5
dataset 3
citation-polarity summary
representative citing papers
DeepMath-103K is a new 103K-problem mathematical dataset with high difficulty, rigorous decontamination, and verifiable answers to support RL training of language-model reasoning.
KCSAT-ML benchmark supplies human error rates for math problems and DRG metric exposes that model accuracy collapses on high-human-error items while test-time scaling shows non-monotonic gains and alignment failures.
ATLAS traces RLVR data to 20 atomic sources, most datasets are variants, and DAPO++ curated with SCA improves RLVR performance while Q predicts training effectiveness.
Frontier LLMs achieve 95-100% accuracy on AMC/AIME problems but recover far fewer distinct valid strategies than human references, while collectively generating 50 novel strategies.
Presents verifiable counterfactual process supervision that generates annotated trajectories via template-aware error injection on symbolic chains, improving Best-of-8 reranking on logical reasoning benchmarks with preliminary math transfer.
UCPO modifies GRPO with a uniformity penalty over correct solutions to prevent diversity collapse in RLVR, yielding up to 10% higher Pass@64 on AIME24 and 45% more equation-level diversity.
OptiVerse is a new benchmark spanning neglected optimization domains that shows LLMs suffer sharp accuracy drops on hard problems due to modeling and logic errors, with a Dual-View Auditor Agent proposed to improve performance.
ProofGrid is a new benchmark for LLM reasoning that uses machine-checkable proofs in minimal formal notation, revealing progress on basic tasks but major gaps in complex combinatorial and synthesis reasoning.
EngiBench shows LLMs accuracy drops with task complexity, degrades under perturbations, and stays below human performance on open-ended engineering problems.
MathArena evaluates over 50 LLMs on 162 fresh competition problems across seven contests, detects contamination in AIME 2024, and reports top models scoring below 40 percent on IMO 2025 proof tasks.
One training example via RLVR boosts LLM math reasoning from 17.6% to 35.7% average across six benchmarks.
OlymMATH is a 350-problem Olympiad math benchmark combining bilingual natural-language evaluation with Lean 4 formal verification to test LLM reasoning.
LCPO trains L1 reasoning models to adhere to prompt-specified CoT lengths, supporting accuracy-compute trade-offs and yielding short reasoning models that outperform larger baselines at matched lengths.
A 400-entry benchmark and protocol shows tool-augmented agents reach 89.5% compilation but only 60.5% consensus faithfulness, with a 29-point gap; elaboration feedback improves validity most but increases unfaithful compiles.
Epi2Diff extracts cognitive episode sequences from LRM reasoning traces and combines them with semantic features to predict human item difficulty, outperforming baselines on four educational datasets.
SkillOpt introduces a controllable text-space optimizer that evolves agent skills via add/delete/replace edits accepted only on strict held-out validation improvement, reporting consistent gains across 52 model-benchmark-harness combinations.
RMA, a multi-agent system with structured memory and iterative feedback loops, solves 8 out of 10 research-level math problems on the new First Proof benchmark and outperforms GPT-5.2R and Aletheia according to expert evaluation.
BET reduces reasoning tokens by about 55% on average while improving performance across benchmarks by learning to short-solve easy queries, fold early on unsolvable ones, and preserve budget for hard solvable queries.
TIDE-Bench is a new benchmark for tool-integrated reasoning that combines diverse tasks, multi-aspect metrics covering answer quality, process reliability, efficiency and cost, plus filtered challenging test sets.
A critique-and-routing controller cast as a finite-horizon MDP with policy-gradient optimization outperforms one-shot routing baselines on reasoning benchmarks while using the strongest agent for under 25% of calls.
A small RL-trained policy for stepwise model routing between LLM sizes improves the accuracy-cost tradeoff on math benchmarks over handcrafted strategies and matches large process reward model methods.
NeWTral is a non-linear weight translation framework using MoE routing that reduces average attack success rate from 70% to 13% on unsafe domain adapters across Llama, Mistral, Qwen, and Gemma models up to 72B while retaining 90% knowledge fidelity.
OLLM models next-token generation as a latent-indexed set of options, enabling up to 70% math reasoning correctness versus 51% baselines and structure-based alignment via a compact latent policy.
citing papers explorer
-
MathNet: a Global Multimodal Benchmark for Mathematical Reasoning and Retrieval
MathNet delivers the largest multilingual Olympiad math dataset and benchmarks where models like Gemini-3.1-Pro reach 78% on solving but embedding models struggle on equivalent problem retrieval, with retrieval augmentation yielding up to 12% gains.
-
DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning
DeepMath-103K is a new 103K-problem mathematical dataset with high difficulty, rigorous decontamination, and verifiable answers to support RL training of language-model reasoning.
-
KCSAT-ML: Probing Reasoning Models with Nationwide-Cohort Human Difficulty
KCSAT-ML benchmark supplies human error rates for math problems and DRG metric exposes that model accuracy collapses on high-human-error items while test-time scaling shows non-monotonic gains and alignment failures.
-
RLVR Datasets and Where to Find Them: Tracing Data Lineage for Better Training Data
ATLAS traces RLVR data to 20 atomic sources, most datasets are variants, and DAPO++ curated with SCA improves RLVR performance while Q predicts training effectiveness.
-
Beyond Accuracy: Evaluating Strategy Diversity in LLM Mathematical Reasoning
Frontier LLMs achieve 95-100% accuracy on AMC/AIME problems but recover far fewer distinct valid strategies than human references, while collectively generating 50 novel strategies.
-
Verifiable Counterfactual Supervision for Process Reward Models
Presents verifiable counterfactual process supervision that generates annotated trajectories via template-aware error injection on symbolic chains, improving Best-of-8 reranking on logical reasoning benchmarks with preliminary math transfer.
-
Uniform-Correct Policy Optimization: Breaking RLVR's Indifference to Diversity
UCPO modifies GRPO with a uniformity penalty over correct solutions to prevent diversity collapse in RLVR, yielding up to 10% higher Pass@64 on AIME24 and 45% more equation-level diversity.
-
OptiVerse: A Comprehensive Benchmark towards Optimization Problem Solving
OptiVerse is a new benchmark spanning neglected optimization domains that shows LLMs suffer sharp accuracy drops on hard problems due to modeling and logic errors, with a Dual-View Auditor Agent proposed to improve performance.
-
Stress-Testing the Reasoning Competence of LLMs With Proofs Under Minimal Formalism
ProofGrid is a new benchmark for LLM reasoning that uses machine-checkable proofs in minimal formal notation, revealing progress on basic tasks but major gaps in complex combinatorial and synthesis reasoning.
-
EngiBench: A Benchmark for Evaluating Large Language Models on Engineering Problem Solving
EngiBench shows LLMs accuracy drops with task complexity, degrades under perturbations, and stays below human performance on open-ended engineering problems.
-
MathArena: Evaluating LLMs on Uncontaminated Math Competitions
MathArena evaluates over 50 LLMs on 162 fresh competition problems across seven contests, detects contamination in AIME 2024, and reports top models scoring below 40 percent on IMO 2025 proof tasks.
-
Reinforcement Learning for Reasoning in Large Language Models with One Training Example
One training example via RLVR boosts LLM math reasoning from 17.6% to 35.7% average across six benchmarks.
-
Challenging the Boundaries of Reasoning: An Olympiad-Level Math Benchmark for Large Language Models
OlymMATH is a 350-problem Olympiad math benchmark combining bilingual natural-language evaluation with Lean 4 formal verification to test LLM reasoning.
-
L1: Controlling How Long A Reasoning Model Thinks With Reinforcement Learning
LCPO trains L1 reasoning models to adhere to prompt-specified CoT lengths, supporting accuracy-compute trade-offs and yielding short reasoning models that outperform larger baselines at matched lengths.
-
Beyond Compilation: Evaluating Faithful Natural-Language-to-Lean Statement Formalization
A 400-entry benchmark and protocol shows tool-augmented agents reach 89.5% compilation but only 60.5% consensus faithfulness, with a 29-point gap; elaboration feedback improves validity most but increases unfaithful compiles.
-
Cognitive Episodes in LLM Reasoning Traces Enable Interpretable Human Item Difficulty Prediction
Epi2Diff extracts cognitive episode sequences from LRM reasoning traces and combines them with semantic features to predict human item difficulty, outperforming baselines on four educational datasets.
-
SkillOpt: Executive Strategy for Self-Evolving Agent Skills
SkillOpt introduces a controllable text-space optimizer that evolves agent skills via add/delete/replace edits accepted only on strict held-out validation improvement, reporting consistent gains across 52 model-benchmark-harness combinations.
-
RMA: an Agentic System for Research-Level Mathematical Problems
RMA, a multi-agent system with structured memory and iterative feedback loops, solves 8 out of 10 research-level math problems on the new First Proof benchmark and outperforms GPT-5.2R and Aletheia according to expert evaluation.
-
Nice Fold or Hero Call: Learning Budget-Efficient Thinking for Adaptive Reasoning
BET reduces reasoning tokens by about 55% on average while improving performance across benchmarks by learning to short-solve easy queries, fold early on unsolvable ones, and preserve budget for hard solvable queries.
-
TIDE-Bench: Task-Aware and Diagnostic Evaluation of Tool-Integrated Reasoning
TIDE-Bench is a new benchmark for tool-integrated reasoning that combines diverse tasks, multi-aspect metrics covering answer quality, process reliability, efficiency and cost, plus filtered challenging test sets.
-
Iterative Critique-and-Routing Controller for Multi-Agent Systems with Heterogeneous LLMs
A critique-and-routing controller cast as a finite-horizon MDP with policy-gradient optimization outperforms one-shot routing baselines on reasoning benchmarks while using the strongest agent for under 25% of calls.
-
Policy-Guided Stepwise Model Routing for Cost-Effective Reasoning
A small RL-trained policy for stepwise model routing between LLM sizes improves the accuracy-cost tradeoff on math benchmarks over handcrafted strategies and matches large process reward model methods.
-
You Snooze, You Lose: Automatic Safety Alignment Restoration through Neural Weight Translation
NeWTral is a non-linear weight translation framework using MoE routing that reduces average attack success rate from 70% to 13% on unsafe domain adapters across Llama, Mistral, Qwen, and Gemma models up to 72B while retaining 90% knowledge fidelity.
-
OLLM: Options-based Large Language Models
OLLM models next-token generation as a latent-indexed set of options, enabling up to 70% math reasoning correctness versus 51% baselines and structure-based alignment via a compact latent policy.
-
HEALing Entropy Collapse: Enhancing Exploration in Few-Shot RLVR via Hybrid-Domain Entropy Dynamics Alignment
HEAL mitigates entropy collapse in few-shot RLVR by selectively adding general-domain data and aligning trajectory-level entropy dynamics, matching full-shot performance with 32 target samples.
-
The Geometric Reasoner: Manifold-Informed Latent Foresight Search for Long-Context Reasoning
TGR performs manifold-informed latent foresight search to boost trajectory coverage in long-context reasoning tasks by up to 13 AUC points with minimal overhead.
-
LLaDA2.0: Scaling Up Diffusion Language Models to 100B
LLaDA2.0 scales discrete diffusion language models to 100B parameters via systematic conversion from autoregressive models using a 3-phase WSD training scheme and releases open-source 16B and 100B MoE variants.
-
MathFlow: Enhancing the Perceptual Flow of MLLMs for Visual Mathematical Problems
MathFlow decouples perception and inference stages in MLLMs for visual math, with a dedicated perception model delivering gains on the FlowVerse benchmark when paired with existing reasoners.
-
LMM-R1: Empowering 3B LMMs with Strong Reasoning Abilities Through Two-Stage Rule-Based RL
A two-stage RL framework first boosts text reasoning in 3B LMMs then adapts it to multimodal inputs, producing modest benchmark gains of 4.5-4.8%.
-
Bridging the Detection-to-Abstention Gap in Reasoning Models under Insufficient Information
JTS trains reasoning models via supervised warm-up and missing-premise RL to make an explicit answerability commitment that triggers early termination on unanswerable inputs, raising Abstention@Detection near saturation.
-
Rethinking Math Reasoning Evaluation: A Robust LLM-as-a-Judge Framework Beyond Symbolic Rigidity
An LLM-as-a-judge evaluation framework for math reasoning outperforms symbolic methods by accurately assessing diverse answer representations and formats.
-
Humanity's Last Exam
Humanity's Last Exam is a new 2,500-question benchmark at the frontier of human knowledge where state-of-the-art LLMs show low accuracy.
-
Artificial Intelligence for Mathematical Reasoning: An Integrated Survey of Language Models, Neuro-symbolic Systems, and Verified Discovery
An integrated survey organizing AI mathematical reasoning into informal, formal, discovery, and technique axes while cataloging benchmarks and assessing failure modes.
- Dual-Cluster Memory Agent: Resolving Multi-Paradigm Ambiguity in Optimization Problem Solving
- Riemann-Bench: A Benchmark for Moonshot Mathematics