KernelPro combines LLM code generation, roofline-guided tool orchestration, and domain-adapted MCTS to produce GPU kernels that outperform prior automated and some hand-tuned baselines on KernelBench and VeOmni workloads.
hub Canonical reference
Large language model-brained gui agents: A survey, 2024a
Canonical reference. 100% of citing Pith papers cite this work as background.
16
Pith papers citing it
Background
100% of classified citations
hub tools
citation-role summary
background 5
citation-polarity summary
roles
background 5polarities
background 5representative citing papers
Introduces OPT* tasks and two training regimes (solver-guided online policy optimization with rank-based reward shaping and search-based offline RL) plus a theoretical link between search success and information extraction per budget unit, showing empirical gains in optimization-like reasoning.
TTT-RTL performs per-design test-time RL on an LLM policy with EDA-derived PPA rewards and an adaptive KL controller, reducing geometric-mean PPA product by 65.1% on RTLLM v2.0 and ADP by 59.4% on an industrial FPU unit.
TAPER regulates LLM branch parallelism by admitting extra branches opportunistically when predicted externality fits slack, delivering 1.48-1.77x higher goodput than eager or fixed-cap baselines on Qwen3-32B while keeping over 95% SLO attainment.
Parallel thinking in LLMs suffers from overscaling where fixed global budgets waste samples; LanBo predicts per-sample budgets from latent states to raise utilization without hurting accuracy.
OSCAR exploits the generative-discriminative gap in LVLMs to build online preference data with MCTS and dual-granularity rewards for DPO-based calibration, claiming SOTA hallucination reduction and improved multimodal performance.
A Lagrangian-relaxation plus imitation-learning pipeline adaptively allocates test-time compute to LLMs, outperforming uniform baselines by up to 12.8% relative accuracy on MATH while staying within a fixed average budget.
Chain-in-Tree cuts token use, model calls, and runtime by 75-85% in LLM tree search on GSM8K and Math500 by using simple branching-necessity checks, with little accuracy loss in most cases.
DeepSearch embeds MCTS into RLVR training with global frontier selection, entropy guidance, and adaptive replay to achieve 62.95% average accuracy on math reasoning benchmarks while using 5.7x fewer GPU hours than extended training.
DITS replaces Q-value guidance in MCTS with influence scores for synthetic data synthesis in multi-agent LLM training, claiming better efficiency and performance on eight datasets.
Introduces consensus objective aggregation for meta-optimization of scientific discovery and reports improved scaling and speedup for 3-SAT algorithm discovery using digital MemComputing machines.
A disagreement-based router that decides per math problem whether to accept, majority-vote, or rewrite improves accuracy by 3–7% with fewer samples than fixed test-time scaling methods.
Bilevel optimization with outer-loop MCTS for skill structure and inner-loop LLM refinement improves agent accuracy on an operations-research question-answering dataset.
A hypothesis-driven pipeline generates targeted hard math problems that drop Llama-3.3-70B-Instruct accuracy from 77% on MATH to as low as 45%.
Graphs can help LLMs reduce hallucinations, boost reasoning via prompting techniques, and better process structured data.
The survey organizes the shift of LLMs toward deliberate System 2 reasoning, covering model construction techniques, performance on math and coding benchmarks, and future research directions.
citing papers explorer
-
Optimizing CUDA like a Human: Micro-Profiling Tools as Expert Surrogates for LLM-Based GPU Kernel Optimization
KernelPro combines LLM code generation, roofline-guided tool orchestration, and domain-adapted MCTS to produce GPU kernels that outperform prior automated and some hand-tuned baselines on KernelBench and VeOmni workloads.
-
Step-by-Step Optimization-like Reasoning in LLMs over Expanding Search Spaces
Introduces OPT* tasks and two training regimes (solver-guided online policy optimization with rank-based reward shaping and search-based offline RL) plus a theoretical link between search success and information extraction per budget unit, showing empirical gains in optimization-like reasoning.
-
Alpha-RTL: Test-Time Training for RTL Hardware Optimization
TTT-RTL performs per-design test-time RL on an LLM policy with EDA-derived PPA rewards and an adaptive KL controller, reducing geometric-mean PPA product by 65.1% on RTLLM v2.0 and ADP by 59.4% on an industrial FPU unit.
-
Regulating Branch Parallelism in LLM Serving
TAPER regulates LLM branch parallelism by admitting extra branches opportunistically when predicted externality fits slack, delivering 1.48-1.77x higher goodput than eager or fixed-cap baselines on Qwen3-32B while keeping over 95% SLO attainment.
-
On the Overscaling Curse of Parallel Thinking: System Efficacy Contradicts Sample Efficiency
Parallel thinking in LLMs suffers from overscaling where fixed global budgets waste samples; LanBo predicts per-sample budgets from latent states to raise utilization without hurting accuracy.
-
Online Self-Calibration Against Hallucination in Vision-Language Models
OSCAR exploits the generative-discriminative gap in LVLMs to build online preference data with MCTS and dual-granularity rewards for DPO-based calibration, claiming SOTA hallucination reduction and improved multimodal performance.
-
Adaptive Test-Time Compute Allocation for Reasoning LLMs via Constrained Policy Optimization
A Lagrangian-relaxation plus imitation-learning pipeline adaptively allocates test-time compute to LLMs, outperforming uniform baselines by up to 12.8% relative accuracy on MATH while staying within a fixed average budget.
-
Chain-in-Tree: Back to Sequential Reasoning in LLM Tree Search
Chain-in-Tree cuts token use, model calls, and runtime by 75-85% in LLM tree search on GSM8K and Math500 by using simple branching-necessity checks, with little accuracy loss in most cases.
-
DeepSearch: Overcome the Bottleneck of Reinforcement Learning with Verifiable Rewards via Monte Carlo Tree Search
DeepSearch embeds MCTS into RLVR training with global frontier selection, entropy guidance, and adaptive replay to achieve 62.95% average accuracy on math reasoning benchmarks while using 5.7x fewer GPU hours than extended training.
-
Efficient Multi-Agent System Training with Data Influence-Oriented Tree Search
DITS replaces Q-value guidance in MCTS with influence scores for synthetic data synthesis in multi-agent LLM training, claiming better efficiency and performance on eight datasets.
-
Scientific discovery as meta-optimization: a combinatorial optimization case study
Introduces consensus objective aggregation for meta-optimization of scientific discovery and reports improved scaling and speedup for 3-SAT algorithm discovery using digital MemComputing machines.
-
When to Vote, When to Rewrite: Disagreement-Guided Strategy Routing for Test-Time Scaling
A disagreement-based router that decides per math problem whether to accept, majority-vote, or rewrite improves accuracy by 3–7% with fewer samples than fixed test-time scaling methods.
-
Bilevel Optimization of Agent Skills via Monte Carlo Tree Search
Bilevel optimization with outer-loop MCTS for skill structure and inner-loop LLM refinement improves agent accuracy on an operations-research question-answering dataset.
-
Automatically Generating Hard Math Problems from Hypothesis-Driven Error Analysis
A hypothesis-driven pipeline generates targeted hard math problems that drop Llama-3.3-70B-Instruct accuracy from 77% on MATH to as low as 45%.
-
Position: How can Graphs Help Large Language Models?
Graphs can help LLMs reduce hallucinations, boost reasoning via prompting techniques, and better process structured data.
-
From System 1 to System 2: A Survey of Reasoning Large Language Models
The survey organizes the shift of LLMs toward deliberate System 2 reasoning, covering model construction techniques, performance on math and coding benchmarks, and future research directions.