Empirical evaluation on LiveCodeBench shows certainty-based RLIF yields early gains followed by output shortening and reasoning collapse, providing no advantage for RLVR initialization on code tasks.
hub Canonical reference
Stepcoder: Improve code generation with reinforcement learning from compiler feedback.arXiv preprint arXiv:2402.01391
Canonical reference. 86% of citing Pith papers cite this work as background.
19
Pith papers citing it
Background
86% of classified citations
hub tools
citation-role summary
background 7
citation-polarity summary
roles
background 7representative citing papers
BoostLoRA grows effective adapter rank linearly via iterative boosting on hard examples with orthogonal low-rank updates, outperforming both single-shot ultra-low-rank adapters and full fine-tuning on math and code tasks with zero added inference overhead.
EyeMulator augments CodeLLM fine-tuning loss with token weights derived from human eye-tracking scan paths, producing large gains on code translation and summarization across StarCoder, Llama-3.2 and DeepSeek-Coder.
The LLM-as-Environment-Engineer framework lets the policy model redesign its own RL environments on the new MAPF-FrozenLake testbed, outperforming larger models and fixed baselines with Qwen3-4B.
CoT SFT disrupts long-range routing in hybrid models via changes to W_Q and W_K; QK-Restore restores pre-SFT projections to recover NIAH performance.
REC RL improves LLM code generation by automatically assessing and optimizing requirement difficulty with adaptive curriculum sampling, yielding 1.23-5.62% Pass@1 gains over baselines.
WebGen-R1 uses end-to-end RL with scaffold-driven generation and cascaded rewards for structure, function, and aesthetics to transform a 7B model into a generator of deployable multi-page websites that rivals much larger models.
AutoOR uses synthetic data generation and RL post-training with solver feedback to enable 8B LLMs to autoformalize linear, mixed-integer, and non-linear OR problems, matching larger models on benchmarks.
MEDS improves LLM RL performance by up to 4.13 pass@1 and 4.37 pass@128 points by dynamically penalizing rollouts matching prevalent historical error clusters identified via memory-stored representations and density clustering.
MR-Adopt deduces input transformations from hard-coded MR test cases using LLMs, data-flow refinement, and output-relation selection to enable reuse with new source inputs.
MAGNIFIED applies RL fine-tuning to MLLMs for autonomous driving motion planning, yielding over 10.5% lower overlap rate and 38.9% lower off-road rate than SFT baseline on Waymo Open Motion Dataset.
ChemDFM-R is a chemical reasoning LLM trained via a four-stage pipeline on the ChemFG dataset of functional-group annotations for molecules and reactions, reaching performance comparable to or better than commercial models on chemical benchmarks.
The paper unifies perspectives on Long CoT in reasoning LLMs by introducing a taxonomy, detailing characteristics of deep reasoning and reflection, and discussing emergence phenomena and future directions.
RLVR with combined unit-test and static-analysis rewards improves pass@1 by up to 13pp on MBPP for 0.6B-1B models, while single-reward variants can induce shorter but less correct outputs.
Offline RL post-training boosts code generation performance in LLMs, with larger gains for small models and hard problems, using pre-collected datasets.
SFT followed by RLVR on Qwen2.5-3B-Instruct raises syntactic and execution correctness when generating Game Code World Models across 30 games.
Static checking rewards and moderate AST-based hints improve diffusion RL performance for code generation, with effectiveness varying by task difficulty across HumanEval, MBPP, and LiveCodeBench.
Proposes autopoietic architectures for self-constructing software as a fundamental shift in the SDLC, leveraging foundation models for autonomous evolution and maintenance.
A systematic literature review that organizes recent work on LLMs for code generation into a taxonomy covering data curation, model advances, evaluations, ethics, environmental impact, and applications, with benchmark comparisons.
citing papers explorer
-
When Do Intrinsic Rewards Work for Code Reasoning? A Comprehensive Study
Empirical evaluation on LiveCodeBench shows certainty-based RLIF yields early gains followed by output shortening and reasoning collapse, providing no advantage for RLVR initialization on code tasks.
-
BoostLoRA: Growing Effective Rank by Boosting Adapters
BoostLoRA grows effective adapter rank linearly via iterative boosting on hard examples with orthogonal low-rank updates, outperforming both single-shot ultra-low-rank adapters and full fine-tuning on math and code tasks with zero added inference overhead.
-
EyeMulator: Improving Code Language Models by Mimicking Human Visual Attention
EyeMulator augments CodeLLM fine-tuning loss with token weights derived from human eye-tracking scan paths, producing large gains on code translation and summarization across StarCoder, Llama-3.2 and DeepSeek-Coder.
-
From Trainee to Trainer: LLM-Designed Training Environment for RL with Multi-Agent Reasoning
The LLM-as-Environment-Engineer framework lets the policy model redesign its own RL environments on the new MAPF-FrozenLake testbed, outperforming larger models and fixed baselines with Qwen3-4B.
-
Attention Amnesia in Hybrid LLMs: When CoT Fine-Tuning Breaks Long-Range Recall, and How to Fix It
CoT SFT disrupts long-range routing in hybrid models via changes to W_Q and W_K; QK-Restore restores pre-SFT projections to recover NIAH performance.
-
Improving LLM Code Generation via Requirement-Aware Curriculum Reinforcement Learning
REC RL improves LLM code generation by automatically assessing and optimizing requirement difficulty with adaptive curriculum sampling, yielding 1.23-5.62% Pass@1 gains over baselines.
-
WebGen-R1: Incentivizing Large Language Models to Generate Functional and Aesthetic Websites with Reinforcement Learning
WebGen-R1 uses end-to-end RL with scaffold-driven generation and cascaded rewards for structure, function, and aesthetics to transform a 7B model into a generator of deployable multi-page websites that rivals much larger models.
-
AutoOR: Scalably Post-training LLMs to Autoformalize Operations Research Problems
AutoOR uses synthetic data generation and RL post-training with solver feedback to enable 8B LLMs to autoformalize linear, mixed-integer, and non-linear OR problems, matching larger models on benchmarks.
-
The Past Is Not Past: Memory-Enhanced Dynamic Reward Shaping
MEDS improves LLM RL performance by up to 4.13 pass@1 and 4.37 pass@128 points by dynamically penalizing rollouts matching prevalent historical error clusters identified via memory-stored representations and density clustering.
-
MR-Adopt: Automatic Deduction of Input Transformation Function for Metamorphic Testing
MR-Adopt deduces input transformations from hard-coded MR test cases using LLMs, data-flow refinement, and output-relation selection to enable reuse with new source inputs.
-
MAGNIFIED: RL Fine-tuning of Multimodal Large Language Models for Motion Planning
MAGNIFIED applies RL fine-tuning to MLLMs for autonomous driving motion planning, yielding over 10.5% lower overlap rate and 38.9% lower off-road rate than SFT baseline on Waymo Open Motion Dataset.
-
ChemDFM-R: A Chemical Reasoning LLM Enhanced with Atomized Chemical Knowledge
ChemDFM-R is a chemical reasoning LLM trained via a four-stage pipeline on the ChemFG dataset of functional-group annotations for molecules and reactions, reaching performance comparable to or better than commercial models on chemical benchmarks.
-
Towards Reasoning Era: A Survey of Long Chain-of-Thought for Reasoning Large Language Models
The paper unifies perspectives on Long CoT in reasoning LLMs by introducing a taxonomy, detailing characteristics of deep reasoning and reflection, and discussing emergence phenomena and future directions.
-
Improving Small Language Models for Code Generation with Reinforcement Learning from Verification Feedback
RLVR with combined unit-test and static-analysis rewards improves pass@1 by up to 13pp on MBPP for 0.6B-1B models, while single-reward variants can induce shorter but less correct outputs.
-
Efficient Post-training of LLMs for Code Generation With Offline Reinforcement Learning
Offline RL post-training boosts code generation performance in LLMs, with larger gains for small models and hard problems, using pre-collected datasets.
-
Distilling Game Code World Model Generation into Lightweight Large Language Models
SFT followed by RLVR on Qwen2.5-3B-Instruct raises syntactic and execution correctness when generating Game Code World Models across 30 games.
-
Beyond Execution: Static-Analysis Rewards and Hint-Conditioned Diffusion RL for Code Generation
Static checking rewards and moderate AST-based hints improve diffusion RL performance for code generation, with effectiveness varying by task difficulty across HumanEval, MBPP, and LiveCodeBench.
-
Towards Enabling An Artificial Self-Construction Software Life-cycle via Autopoietic Architectures
Proposes autopoietic architectures for self-constructing software as a fundamental shift in the SDLC, leveraging foundation models for autonomous evolution and maintenance.
-
A Survey on Large Language Models for Code Generation
A systematic literature review that organizes recent work on LLMs for code generation into a taxonomy covering data curation, model advances, evaluations, ethics, environmental impact, and applications, with benchmark comparisons.