相关论文: STOP: Structured On-Policy Pruning of Long-Form Re…

Deploying accurate Text-to-SQL systems at the enterprise level faces a difficult trilemma involving cost, security and performance. Current solutions force enterprises to choose between expensive, proprietary Large Language Models (LLMs)…

Multi-hop retrieval-augmented generation (RAG) is a promising strategy for complex reasoning, yet existing iterative prompting approaches remain inefficient. They often regenerate predictable token sequences at every step and rely on…

Reasoning large language models (LLMs) heavily rely on scaling test-time compute to perform complex reasoning tasks by generating extensive "thinking" chains. While demonstrating impressive results, this approach incurs significant…

Recent advances in Large Reasoning Models (LRMs) have demonstrated strong performance on complex tasks through long Chain-of-Thought (CoT) reasoning. However, their lengthy outputs increase computational costs and may lead to overthinking,…

Structured pruning is an effective approach for compressing large pre-trained neural networks without significantly affecting their performance. However, most current structured pruning methods do not provide any performance guarantees, and…

Large reasoning models (LRMs) excel at complex reasoning tasks but typically generate lengthy sequential chains-of-thought, resulting in long inference times before arriving at the final answer. To address this challenge, we introduce…

Large reasoning models (LRMs) have achieved remarkable progress on complex tasks by generating extended chains of thought (CoT). However, their uncontrolled output lengths pose significant challenges for real-world deployment, where…

Chain-of-thought (CoT) distillation aims to enhance small language models' (SLMs) reasoning by transferring multi-step reasoning capability from the larger teacher models. However, existing work underestimates rationale quality, focusing…

Effective relevance modeling is crucial for e-commerce search, as it aligns search results with user intent and enhances customer experience. Recent work has leveraged large language models (LLMs) to address the limitations of traditional…

Reasoning training incentivizes LLMs to produce long chains of thought (long CoT), which among other things, allows them to explore solution strategies with self-checking. This results in higher accuracy, but inflates context length,…

Continual supervised fine-tuning (SFT) is the de facto recipe for adapting large language models (LLMs) to a stream of downstream tasks, but it suffers from catastrophic forgetting of earlier capabilities. Recent work shows that on-policy…

Latent tokens are gaining attention for enhancing reasoning in large language models (LLMs), yet their internal mechanisms remain unclear. This paper examines the problem from a reliability perspective, uncovering fundamental weaknesses:…

Large Reasoning Models (LRMs) often suffer from overthinking, generating verbose reasoning traces that compromise both computational efficiency and interpretability. Unlike prior efforts that rely on global length-based rewards, we propose…

Deploying pre-trained transformer models like BERT on downstream tasks in resource-constrained scenarios is challenging due to their high inference cost, which grows rapidly with input sequence length. In this work, we propose a…

Off-policy deep reinforcement learning (RL) typically leverages replay buffers for reusing past experiences during learning. This can help improve sample efficiency when the collected data is informative and aligned with the learning…

Large Language Models (LLMs) now exhibit remarkable reasoning capabilities through test-time compute scaling (TTS), with impressive performance across math and coding benchmarks. In parallel, research in model compression has developed…

Recent years have witnessed significant progress in large language models' (LLMs) reasoning, which is largely due to the chain-of-thought (CoT) approaches, allowing models to generate intermediate reasoning steps before reaching the final…

Scaling inference compute enhances reasoning in large language models (LLMs), with long chains-of-thought (CoTs) enabling strategies like backtracking and error correction. Reinforcement learning (RL) has emerged as a crucial method for…

Long chain-of-thought (CoT) prompting helps Large Language Models (LLMs) solve difficult problems, but very long traces often slow or even degrade performance on fast, intuitive "System-1" tasks. We introduce Connector-Aware Compact CoT…

Recent studies have shown that making a model spend more time thinking through longer Chain of Thoughts (CoTs) enables it to gain significant improvements in complex reasoning tasks. While current researches continue to explore the benefits…