AI model builders mostly highlight unique benchmarks that act as flexible narrative tools for market positioning rather than standardized scientific measurements.
super hub Baseline reference
LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code
Baseline reference. 55% of citing Pith papers use this work as a benchmark or comparison.
259
Pith papers citing it
Baseline
55% of classified citations
abstract
Large Language Models (LLMs) applied to code-related applications have emerged as a prominent field, attracting significant interest from both academia and industry. However, as new and improved LLMs are developed, existing evaluation benchmarks (e.g., HumanEval, MBPP) are no longer sufficient for assessing their capabilities. In this work, we propose LiveCodeBench, a comprehensive and contamination-free evaluation of LLMs for code, which continuously collects new problems over time from contests across three competition platforms, namely LeetCode, AtCoder, and CodeForces. Notably, our benchmark also focuses on a broader range of code related capabilities, such as self-repair, code execution, and test output prediction, beyond just code generation. Currently, LiveCodeBench hosts four hundred high-quality coding problems that were published between May 2023 and May 2024. We have evaluated 18 base LLMs and 34 instruction-tuned LLMs on LiveCodeBench. We present empirical findings on contamination, holistic performance comparisons, potential overfitting in existing benchmarks as well as individual model comparisons. We will release all prompts and model completions for further community analysis, along with a general toolkit for adding new scenarios and model
hub tools
citation-role summary
dataset 25
background 16
baseline 1
contradiction 1
method 1
citation-polarity summary
claims ledger
- abstract Large Language Models (LLMs) applied to code-related applications have emerged as a prominent field, attracting significant interest from both academia and industry. However, as new and improved LLMs are developed, existing evaluation benchmarks (e.g., HumanEval, MBPP) are no longer sufficient for assessing their capabilities. In this work, we propose LiveCodeBench, a comprehensive and contamination-free evaluation of LLMs for code, which continuously collects new problems over time from contests across three competition platforms, namely LeetCode, AtCoder, and CodeForces. Notably, our benchma
authors
co-cited works
representative citing papers
FlowCompile performs compile-time design space exploration on structured LLM workflows to produce reusable high-quality configuration sets that outperform routing baselines with up to 6.4x speedup.
A new native-runtime benchmark reveals that current frontier AI agents succeed on at most 62 percent of realistic long-horizon CLI tasks.
LiveBench is a contamination-limited LLM benchmark with auto-scored challenging tasks from recent sources across math, coding, reasoning and more, where top models score below 70%.
TestEvo-Bench supplies 746 test-generation and 509 test-update tasks from 152 Java repositories, each tied to actual commits and packaged for execution-based scoring, with current agents reaching 77.5% and 74.6% success respectively.
DecompRL is an RL method that learns modular code decomposition for LLMs, enabling exponential candidate generation via recombination to solve harder coding problems with lower GPU cost.
AxDafny achieves 92.7% verification success on DafnyBench (6.5 points above prior proof-hint baselines) via verifier-guided repair and introduces the LCB-Pro-Dafny benchmark of 250 problems.
AlgoBench creates traceable variants of competitive programming problems via constraint shifts that invalidate original algorithms, paired with complexity metrics that reveal LLMs often produce functionally correct but asymptotically unsuitable solutions.
Introduces the Generalization Spectrum evaluation framework to track per-example generalization across transfer distances in competitive programming tasks.
RigorBench evaluates AI coding agents on process discipline via five pillars and reports 41% higher process scores and 17% better outcome correctness with structured approaches on 30 tasks.
Introduces the Power Systems Agent Benchmark with 41 task families across eight power engineering areas for executable evaluation of AI agents using deterministic feasibility checks.
BIM-Edit benchmark finds best LLM scores only 49.5% average across geometric, semantic, and topological metrics on 324 IFC editing tasks, with no model fully solving more than 3.4%.
A new code-writing data analysis benchmark shows human experts outperforming a frontier LLM on average with lower performance variance.
The paper introduces Uni-E, a unified energy for DLMs that accounts for model capacity, dependency and invariance, can be computed exactly, and corrects distribution shifts from dependency and invariance.
PRISM is a contrastive, policy-aware training framework for process reward models that reduces false positives by 22% on PRMBench and boosts downstream accuracy up to 33% in Best-of-N selection by learning reliable relative comparisons instead of pointwise labels.
OPD updates occupy a relaxed off-principal regime and rapidly lock into a low-dimensional subspace that is functionally sufficient for its performance, distinct from SFT and RLVR trajectories.
SkelDPO improves code generation efficiency by 2-7% over prior DPO methods via joint preference losses on full code and efficiency-critical skeletons.
DistIL applies distributional DAgger with forward cross-entropy to achieve monotonic policy improvement and better Pass@N from rich feedback in RL for reasoning tasks.
CoEval generates task-specific benchmarks by rotating models through teacher, student, and judge roles, then weights questions by discriminative power and judges by panel consensus to recover accurate model rankings without labels.
ResMerge improves merging of RL expert LLMs via a stable residual consensus backbone plus gated head correction, outperforming task-vector and spectral baselines in capability preservation.
ATLAS introduces an LLM-orchestrated agentic framework for dynamic test-time scaling via extensible 'explore' actions, achieving higher accuracy with fewer API calls than fixed-workflow baselines on four benchmarks.
BASTION is a budget-aware speculative decoding framework with adaptive tree-structured block diffusion drafting that reports up to 6.61x speedup and 39% improvement over block-diffusion baselines.
RepoMirage uses semantics-preserving perturbations on SWE-Bench to show code agents lack repository context reasoning, with performance falling sharply on extended structure tasks, and introduces RepoAnchor as a structure-first fix.
MCPO applies contrastive learning to GRPO-style RL by treating cross-domain correct rollouts as positives and incorrect ones as negatives to improve multi-domain reasoning performance in LRMs.
citing papers explorer
-
Think Anywhere in Code Generation
Think-Anywhere lets LLMs invoke on-demand reasoning at any token during code generation via cold-start imitation followed by outcome-based RL, reaching state-of-the-art results on LeetCode, LiveCodeBench, HumanEval, and MBPP.
-
You Don't Need Public Tests to Generate Correct Code
DryRUN lets LLMs create their own test inputs and run internal simulations for self-correcting code generation, matching the performance of test-dependent methods like CodeSIM on LiveCodeBench without public tests or external signals.
-
Co-Located Tests, Better AI Code: How Test Syntax Structure Affects Foundation Model Code Generation
Co-locating tests with implementation code yields substantially higher preservation and correctness in foundation-model-generated programs than separated test syntax.
-
Generating Verifiable Chain of Thoughts from Exection-Traces
A pipeline produces 54,000 execution-trace-verified bi-directional Chain-of-Thought rationales for code, and fine-tuning on them yields gains up to 26.6 points on LiveCodeBench-Exec and similar benchmarks.