Natural Language Processing

O1-Pruner efficiently prunes long-thought reasoning in LLMs by harmonizing reasoning length and accuracy via fine-tuning, significantly reducing inference time without sacrificing performance.

Kimi K1.5: A Multimodal LLM trained with RL achieves state-of-the-art reasoning by scaling long context RL training and improving policy optimization.

DeepSeek-R1 significantly improves LLM reasoning by using reinforcement learning, achieving performance comparable to OpenAI’s top models while addressing previous challenges of poor readability and l…

Revolutionizing large language models, Autonomy-of-Experts (AoE) empowers individual expert modules to autonomously select inputs, eliminating routers and boosting both efficiency and accuracy.

MMVU: a new benchmark pushes multimodal video understanding to expert level, revealing limitations of current models and paving the way for more advanced AI.

Debate-enhanced weak supervision boosts AI alignment by combining strong and weak models, enabling safer and more reliable AI systems.

This research proposes principles and a framework to tackle redundancy in MLLM benchmarks, enhancing efficiency and guiding future development.

Democratizing advanced reasoning in AI, this blueprint introduces a modular framework for building Reasoning Language Models (RLMs), simplifying development and enhancing accessibility.

Mobile-Agent-E: A self-evolving mobile assistant conquering complex tasks with hierarchical agents and a novel self-evolution module, significantly outperforming prior approaches.

Agent-R: A novel self-training framework enables language model agents to learn from errors by dynamically constructing training data that corrects erroneous actions, resulting in significantly improv…

IntellAgent: a novel open-source framework automating diverse conversational AI evaluation via policy-driven graph modeling, event generation, and user-agent simulations, enabling fine-grained diagnos…

STEP-KTO: A novel training framework boosts LLMs’ mathematical reasoning by providing binary feedback on both intermediate steps and final answers. This ensures logical reasoning trajectories and impr…

PaSa: An LLM agent autonomously performs comprehensive academic paper searches, outperforming existing methods by efficiently combining search tools, paper reading, and citation analysis, optimized vi…

Mind Evolution, a novel evolutionary search strategy, significantly boosts Large Language Model (LLM) problem-solving by generating, recombining, and refining candidate solutions via an LLM, outperfor…

ComplexFuncBench, a new benchmark, rigorously evaluates LLMs’ complex function-calling abilities across real-world scenarios involving multi-step processes, constraints, and long contexts.

This survey paper explores the exciting new frontier of Large Reasoning Models (LRMs), focusing on how reinforcement learning and clever prompting techniques are boosting LLMs’ reasoning capabilities.

LLM reasoning boosts self-confidence, even when answers are wrong, highlighting limitations in current evaluation metrics.

AI-powered medical consultations often struggle with the inquiry phase. This paper presents a novel patient simulator trained on real interactions, revealing that effective inquiry significantly impac…

Arabic LLMs struggle with medical tasks; this study reveals optimal language ratios in training data for improved performance, highlighting challenges in simply translating medical data for different …

RLHS, a novel alignment algorithm, leverages simulated hindsight feedback to mitigate misalignment in RLHF, significantly improving AI’s alignment with human values and goals.