中文
相关论文

相关论文: PonderLM-2: Pretraining LLM with Latent Thoughts i…

200 篇论文

Chain-of-Thought (CoT) prompting has significantly advanced task-solving capabilities in natural language processing with large language models. Unlike standard prompting, CoT encourages the model to generate intermediate reasoning steps,…

Chain of thought finetuning (cot-finetuning) aims to endow small language models (SLM) with reasoning ability to improve their performance towards specific tasks by allowing them to imitate the reasoning procedure of large language models…

计算与语言 · 计算机科学 2025-08-08 Xiaoshu Chen , Sihang Zhou , Ke Liang , Xinwang Liu

Reasoning-oriented Large Language Models (LLMs) often rely on generating explicit tokens step by step, and their effectiveness typically hinges on large-scale supervised fine-tuning or reinforcement learning. While Chain-of-Thought (CoT)…

We investigate whether the success of a zero-shot Chain-of-Thought (CoT) process can be predicted before completion. We discover that a probing classifier, based on LLM representations, performs well \emph{even before a single token is…

计算与语言 · 计算机科学 2025-06-03 Anum Afzal , Florian Matthes , Gal Chechik , Yftah Ziser

Continuous chain-of-thought has been shown to be effective in saving reasoning tokens for large language models. By reasoning with continuous latent thought tokens, continuous CoT is able to perform implicit reasoning in a compact manner.…

计算与语言 · 计算机科学 2026-02-27 Haoyi Wu , Zhihao Teng , Kewei Tu

Large Language Models (LLMs) achieve superior performance through Chain-of-Thought (CoT) reasoning, but these token-level reasoning chains are computationally expensive and inefficient. In this paper, we introduce Compressed Latent…

计算与语言 · 计算机科学 2026-02-04 Wenhui Tan , Jiaze Li , Jianzhong Ju , Zhenbo Luo , Ruihua Song , Jian Luan

In-context learning with large language models (LLMs) has recently caught increasing attention due to its superior few-shot performance on various tasks. However, its performance on text-to-SQL parsing still has much room for improvement.…

计算与语言 · 计算机科学 2023-10-30 Chang-You Tai , Ziru Chen , Tianshu Zhang , Xiang Deng , Huan Sun

Large Language Models, such as Generative Pre-trained Transformer 3 (aka. GPT-3), have been developed to understand language through the analysis of extensive text data, allowing them to identify patterns and connections between words.…

计算与语言 · 计算机科学 2023-10-03 Baphumelele Masikisiki , Vukosi Marivate , Yvette Hlope

Large Language Models (LLMs) excel at problem solving by generating chain of thoughts in natural language, but such verbal thinking is computationally costly and prone to overthinking. A recent work instead proposes a latent thinking…

计算与语言 · 计算机科学 2026-02-25 Hanwen Du , Yuxin Dong , Xia Ning

Recent advancements in large language models (LLMs) have demonstrated that progressive refinement, rather than providing a single answer, results in more accurate and thoughtful outputs. However, existing methods often rely heavily on…

Despite superior reasoning prowess demonstrated by Large Language Models (LLMs) with Chain-of-Thought (CoT) prompting, a lack of understanding prevails around the internal mechanisms of the models that facilitate CoT generation. This work…

计算与语言 · 计算机科学 2024-05-07 Subhabrata Dutta , Joykirat Singh , Soumen Chakrabarti , Tanmoy Chakraborty

Modern language models generate chain-of-thought traces by autoregressively sampling tokens from a finite vocabulary. While this discrete sampling has achieved remarkable success, conducting chain-of-thought with continuously-valued tokens…

Pre-trained language models (LMs) have shown remarkable reasoning performance using explanations or chain-of-thoughts (CoT)) for in-context learning. On the other hand, these reasoning tasks are usually presumed to be more approachable for…

计算与语言 · 计算机科学 2024-03-29 Yi-Fan Zhang , Hanlin Zhang , Li Erran Li , Eric Xing

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:…

计算与语言 · 计算机科学 2025-12-29 Yuyi Zhang , Boyu Tang , Tianjie Ju , Sufeng Duan , Gongshen Liu

Current approaches for scaling inference-time compute in transformers train them to emit explicit chain-of-thought tokens before producing an answer. While these methods are powerful, they are limited because they cannot be applied during…

机器学习 · 计算机科学 2026-02-02 Houjun Liu , Shikhar Murty , Christopher D. Manning , Róbert Csordás

Chain-of-Thought (CoT) significantly enhances formal reasoning capabilities in Large Language Models (LLMs) by training them to explicitly generate intermediate reasoning steps. While LLMs readily benefit from such techniques, improving…

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,…

While Long Chain-of-Thought (CoT) reasoning significantly improves Large Language Models (LLMs) performance on complex reasoning tasks, the substantial computational and memory costs of generating long CoT sequences limit their efficiency…

人工智能 · 计算机科学 2026-02-03 Liang Zhang , Yu Zhao , Longyue Wang , Tianqi Shi , Weihua Luo , Kaifu Zhang , Jinsong Su

Large language models (LLMs) excel at complex tasks thanks to advances in their reasoning abilities. However, existing methods overlook the trade-off between reasoning effectiveness and efficiency, often encouraging unnecessarily long…

机器学习 · 计算机科学 2025-10-16 Jingyao Wang , Wenwen Qiang , Zeen Song , Changwen Zheng , Hui Xiong

Large language models (LLMs) can perform complex reasoning by generating intermediate reasoning steps. Providing these steps for prompting demonstrations is called chain-of-thought (CoT) prompting. CoT prompting has two major paradigms. One…

计算与语言 · 计算机科学 2022-10-10 Zhuosheng Zhang , Aston Zhang , Mu Li , Alex Smola