DuST self-trains LLMs for code generation by ranking their own test-time samples via sandbox execution and applying GRPO, improving judgment by +6.2 NDCG and single-sample pass@1 by +3.1 on LiveCodeBench.
hub
S*: Test time scaling for code generation
16 Pith papers cite this work. Polarity classification is still indexing.
16
Pith papers citing it
hub tools
citation-role summary
background 3
dataset 1
citation-polarity summary
representative citing papers
Clover fixes 96.8% of bugs on an RTL-repair benchmark using stochastic tree-of-thoughts and neural-symbolic agents, outperforming traditional and LLM baselines by 94% and 63% respectively with 87.5% pass@1.
AdverMCTS frames code generation as a minimax game where an attacker evolves tests to expose flaws in solver-generated code, yielding more robust outputs than static-test baselines.
A new dataset and nine-metric majority-vote procedure show that existing code-reasoning benchmarks are dominated by lower-complexity problems that do not reflect real-world code.
The first empirical study of test overfitting shows that auto-generated tests from issues can lead to code that passes observed tests but misses important cases or breaks functionality in SWE-bench issue resolution.
E-TTS introduces a plug-and-play test-time scaling method for embodied tasks that unifies reasoning-action sampling with history buffers and closed-loop refinement to improve performance on manipulation benchmarks.
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.
RL-trained lightweight controller using answer statistics improves trade-offs among correctness, latency, and total samples in adaptive sampling for LLM test-time scaling.
Proposes PDF, a hierarchical multi-agent Perception-to-Deliberation Framework that adds experience self-evolution and test-time scaling to composed image retrieval, claiming SOTA on CIRR, CIRCO, and FashionIQ.
VLA-ATTC equips VLA models with adaptive test-time compute via an uncertainty clutch and relative action critic, cutting failure rates by over 50% on LIBERO-LONG.
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.
SPS interleaves RL and IRL to counteract probability squeezing in LLM reasoning trajectories, improving Pass@k on five benchmarks while identifying an empirical upper bound on multi-sample performance.
Ensemble Semantic Entropy improves correlation with code correctness over single-model methods and powers a cascading scaling system that cuts FLOPs by 64.9% while preserving performance on LiveCodeBench.
Lack of exploration from conditioning on prior answers is the primary reason parallel sampling outperforms sequential sampling in large reasoning models.
citing papers explorer
-
Primal Generation, Dual Judgment: Self-Training from Test-Time Scaling
DuST self-trains LLMs for code generation by ranking their own test-time samples via sandbox execution and applying GRPO, improving judgment by +6.2 NDCG and single-sample pass@1 by +3.1 on LiveCodeBench.
-
Clover: A Neural-Symbolic Agentic Harness with Stochastic Tree-of-Thoughts for Verified RTL Repair
Clover fixes 96.8% of bugs on an RTL-repair benchmark using stochastic tree-of-thoughts and neural-symbolic agents, outperforming traditional and LLM baselines by 94% and 63% respectively with 87.5% pass@1.
-
AdverMCTS: Combating Pseudo-Correctness in Code Generation via Adversarial Monte Carlo Tree Search
AdverMCTS frames code generation as a minimax game where an attacker evolves tests to expose flaws in solver-generated code, yielding more robust outputs than static-test baselines.
-
Evaluating Code Reasoning Abilities of Large Language Models Under Real-World Settings
A new dataset and nine-metric majority-vote procedure show that existing code-reasoning benchmarks are dominated by lower-complexity problems that do not reflect real-world code.
-
Investigating Test Overfitting on SWE-bench
The first empirical study of test overfitting shows that auto-generated tests from issues can lead to code that passes observed tests but misses important cases or breaks functionality in SWE-bench issue resolution.
-
E-TTS: A New Embodied Test-Time Scaling Framework for Robotic Manipulation
E-TTS introduces a plug-and-play test-time scaling method for embodied tasks that unifies reasoning-action sampling with history buffers and closed-loop refinement to improve performance on manipulation benchmarks.
-
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.
-
Small RL Controller, Large Language Model: RL-Guided Adaptive Sampling for Test-Time Scaling
RL-trained lightweight controller using answer statistics improves trade-offs among correctness, latency, and total samples in adaptive sampling for LLM test-time scaling.
-
DeliCIR: Deliberative Test-Time Evolutionary Hierarchical Multi-Agents for Composed Image Retrieval
Proposes PDF, a hierarchical multi-agent Perception-to-Deliberation Framework that adds experience self-evolution and test-time scaling to composed image retrieval, claiming SOTA on CIRR, CIRCO, and FashionIQ.
-
VLA-ATTC: Adaptive Test-Time Compute for VLA Models with Relative Action Critic Model
VLA-ATTC equips VLA models with adaptive test-time compute via an uncertainty clutch and relative action critic, cutting failure rates by over 50% on LIBERO-LONG.
-
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.
-
SPS: Steering Probability Squeezing for Better Exploration in Reinforcement Learning for Large Language Models
SPS interleaves RL and IRL to counteract probability squeezing in LLM reasoning trajectories, improving Pass@k on five benchmarks while identifying an empirical upper bound on multi-sample performance.
-
Ensemble-Based Uncertainty Estimation for Code Correctness Estimation
Ensemble Semantic Entropy improves correlation with code correctness over single-model methods and powers a cascading scaling system that cuts FLOPs by 64.9% while preserving performance on LiveCodeBench.
-
Understanding Performance Gap Between Parallel and Sequential Sampling in Large Reasoning Models
Lack of exploration from conditioning on prior answers is the primary reason parallel sampling outperforms sequential sampling in large reasoning models.
- An Iterative Test-and-Repair Framework for Competitive Code Generation