5arXiv cs.AI (Artificial Intelligence)·
MineValiCoder: Bipartite graph mutual validation improves LLM-based test-driven code generation
MineValiCoder is a closed-loop test-driven development framework that addresses LLM stochasticity in automated code generation by combining test-case quality mining, parallel TDD refinement, and bipartite graph-based code-test mutual validation. The system filters faulty auto-generated test cases and uses validated feedback to iteratively optimize code candidates before selecting the best via mutual validation scoring. Evaluated across four LLMs, it achieves 96.34% Pass@1 on HumanEval, 87.40% on MBPP, 64.00% on APPS, and 51.33% on LiveCodeBench, outperforming prior state-of-the-art methods.
Related guides (2)
Related events (8)
PowerCodeBench: Knowledge Boundary Probing and Intervention for LLM-Based Power System Code Generation
This paper introduces PowerCodeBench, an execution-validated benchmark for evaluating LLMs on power-system simulation code generation using the pandapower library. The authors identify that failures are dominated by API-knowledge boundary errors (hallucinated function names, misused parameters) rather than reasoning failures, and propose a boundary-aware intervention combining API demand estimation with targeted documentation injection. Evaluated across ten open-weight models (1.5B–480B) and four commercial APIs on 2,000 tasks, the intervention yields 32–56 accuracy point improvements while using only 41% of baseline prompt-token cost. Open-weight models in the 70B–120B range match commercial mid-tier accuracy, with Llama-3.1-405B and Qwen3-Coder-480B leading.
LLM-as-a-Verifier: Training-free verification framework scales along granularity, repetition, and criteria decomposition
Researchers introduce LLM-as-a-Verifier, a general-purpose verification framework that treats verification as a new scaling axis for LLMs, computing continuous scores from token logit distributions rather than discrete judge outputs. The framework scales along three dimensions—score granularity, repeated evaluation, and criteria decomposition—and achieves state-of-the-art results on Terminal-Bench V2 (86.5%), SWE-Bench Verified (78.2%), RoboRewardBench (87.4%), and MedAgentBench (73.3%) without requiring additional training. The authors also demonstrate that the framework's fine-grained signals can serve as dense RL feedback, improving sample efficiency for SAC and GRPO on robotics and math benchmarks, and build a Claude Code extension for monitoring agentic systems.
Introducing the LiveCodeBench Leaderboard - Holistic and Contamination-Free Evaluation of Code LLMs
Hugging Face introduces a leaderboard based on LiveCodeBench, a benchmark designed for holistic and contamination-free evaluation of code-generating large language models. The benchmark continuously collects new coding problems from competitive programming platforms to prevent data contamination that plagues static benchmarks. It evaluates models across multiple code-related tasks beyond just code generation, aiming to provide a more reliable signal of true model capability.
Generative Compilation: Real-time compiler feedback during LLM code generation via sealors
Researchers introduce generative compilation, a technique that provides compiler feedback on partial programs during autoregressive LLM decoding rather than only after generation completes. The core mechanism is a 'sealor'—a syntax-guided transformation that converts partial programs into complete ones that standard compilers can analyze, formally verified in Lean. Evaluated on repository-level Rust coding tasks across frontier black-box and open-weight models, the approach reduces non-compiling outputs and improves functional correctness by catching errors early and preventing error cascades. The method works without white-box model access, distinguishing it from constrained decoding approaches.
BINEVAL: Binary question decomposition for interpretable LLM evaluation and prompt optimization
Researchers introduce BINEVAL, a framework that decomposes LLM evaluation criteria into atomic binary yes/no questions, aggregating answers into multi-dimensional interpretable scores. The approach matches or outperforms baselines including UniEval and G-Eval on SummEval, Topical-Chat, and QAGS benchmarks, with particular strength on factual consistency. Beyond evaluation, the binary question feedback is shown to support iterative prompt optimization in both self-update and cross-model settings on IFBench. The framework is training-free and task-agnostic, addressing opacity and ceiling-effect problems common in holistic LLM judges.
LLMs judge worse than they generate: empirical challenge to self-evaluation pipeline assumptions
A new arXiv preprint tests the implicit assumption that LLM evaluation is easier than generation, using a controlled in-context QA setup across four benchmarks (SQuAD 2.0, DROP, HotpotQA, MuSiQue) and two models. Results show generation accuracy exceeds self-evaluation accuracy on three of four benchmarks, with attention analysis revealing that evaluation attends to context 3–5x less than generation does. LoRA fine-tuning experiments confirm the asymmetry is not a training artifact, with cross-task interference observed in both directions. The findings directly challenge assumptions underlying LLM-as-a-Judge and self-evaluation pipelines widely used in RLHF and agentic systems.
AdversaBench: Automated LLM red-teaming pipeline with multi-judge confirmation and cross-model transferability
AdversaBench is a new end-to-end red-teaming pipeline that mutates seed prompts using five structured operators and confirms failures via a three-judge panel with a meta-judge tiebreaker. Experiments on 45 seeds across reasoning, instruction-following, and tool-use categories produced confirmed failures for every seed. Key findings include sharp variation in operator effectiveness by category, misleading binary failure rates, judge agreement metrics distorted by label skew, and zero-shot transferability of adversarial prompts from Llama 3.1 8B to Llama 3.3 70B. Code and dataset are publicly released.
Double Ratchet: Co-evolving evaluation metrics and skills for self-improving LLM agents
A new arXiv preprint introduces Double Ratchet, a system that co-evolves both evaluation metrics and agent skills in settings where no reliable automatic verifier exists. The metric loop uses evolutionary search over small drawback detectors anchored to a small reference set, while the skill loop uses a lifecycle-managed approach; together they retain 88–110% of the performance lift achievable with ground-truth metrics across code generation (MBPP+), text-to-SQL (Spider 2.0-Snow), and report generation tasks. The paper also addresses safety, showing that anchor discipline and outer audits can catch and repair cases where evolved skills game the rubric. This work directly addresses a core bottleneck in self-improving agent systems: the chicken-and-egg problem of needing a reliable evaluator to improve.