A comprehensive index of artificial intelligence and machine-learning research with AI-generated summaries, citation metrics, and direct links to papers and code.
Bo Wen, Xin Zhang
SOLOMON is a neuro-inspired LLM reasoning network architecture that enhances domain adaptability via prompt engineering and in-context learning, outperforming baselines and matching o1-preview in semiconductor layout design.
Komal Kumar, Tajamul Ashraf, Omkar Thawakar, et al.
A systematic survey of post-training techniques for LLMs, covering fine-tuning, reinforcement learning, and test-time scaling, with a public repository for tracking developments.
Karthik Valmeekam, Kaya Stechly, Subbarao Kambhampati
This paper evaluates current LLMs and new Large Reasoning Models on PlanBench, finding that while OpenAI's o1 shows quantum improvement, it still falls far short of saturating the benchmark.
Arian Hosseini, Alessandro Sordoni, Daniel Toyama, et al.
This paper reveals a significant reasoning gap in LLMs when solving compositional math problems, showing that performance on standard benchmarks masks systematic differences in reasoning abilities.
Siwei Wu, Z. Peng, Xinrun Du, et al.
This paper analyzes OpenAI's o1 model's reasoning patterns, comparing it with existing test-time compute methods on general reasoning benchmarks and summarizing six distinct reasoning patterns.
Daman Arora, Andrea Zanette
This paper proposes using reinforcement learning to train large reasoning models to dynamically allocate inference-time compute based on task complexity, reducing inference costs while preserving accuracy.
Guanghao Ye, Khiem Pham, Xinzhi Zhang, et al.
Proposes RLSP, a post-training framework that decouples exploration and correctness signals during PPO to enable emergent reasoning behaviors in LLMs.
Yang Sui, Yu-Neng Chuang, Guanchu Wang, et al.
This survey systematically categorizes and reviews methods for efficient reasoning in LLMs, addressing the overthinking phenomenon by optimizing models, outputs, and input prompts.
Zhuoshi Pan, Qizhi Pei, Yu Li, et al.
Introduces REST, a stress-testing framework that evaluates large reasoning models by presenting multiple problems simultaneously, revealing performance degradation and stronger discriminative power than single-question benchmarks.
Ryan Sze-Yin Chan, Federico Nanni, Tomas Lazauskas, et al.
This paper presents a retrieval-augmented reasoning system using a lean Qwen2.5-Instruct model fine-tuned with synthetic reasoning traces, achieving near-frontier performance on domain-specific queries while remaining locally deployable.
L. Ruis, Maximilian Mozes, Juhan Bae, et al.
This paper shows that LLMs use procedural knowledge from pretraining documents to solve reasoning tasks, rather than retrieving answers, indicating generalizable strategies.
Arda Sevinc, Abdurrahman Gumus
AutoReason automatically generates chain-of-thought rationales by decomposing implicit queries into explicit questions, improving multi-step reasoning accuracy in weaker LLMs on StrategyQA and HotpotQA.