Reasoning

We propose Prism-Δ, a differential subspace steering method for prompt highlighting, together with Prism-ΔV, which extends it from the routing (Key) channel to the content (Value) channel — matching or exceeding the best existing method on 19 of 20 configurations while halving the fluency cost of steering.

Large Language Models (LLMs) have made remarkable breakthroughs in reasoning, yet continue to struggle with hallucinations, logical …

Reinforcement learning (RL) has shown great promise in enhancing LLM reasoning, but current approaches mainly focus on single domains …

We present PromptCD, a test-time method for controlling LLM behavior without additional training. The approach creates paired positive …

The advancement of large language models (LLMs) has catalyzed a paradigm shift from code generation assistance to autonomous coding …

Vision-Language Models (VLMs) have demonstrated remarkable success across diverse visual tasks, yet their performance degrades in …

We present the first comprehensive analysis of how the order of graph descriptions impacts LLM performance, evaluating four graph description orders across six graph problems using six mainstream LLMs.

We present the first comprehensive analysis of the impacts of CoT prompting on Reasoning LLMs, finding that one-shot CoT consistently enhances performance and reduces excessive reflections by approximately 90%.

The authors propose CoT-Influx, a method that optimizes both the exemplars and the content of chain-of-thought prompts in order to improve the reasoning ability of LLMs. Its core idea is to maximize the amount of useful information fed into the model through pruning.

This paper proposes a new method that jointly assesses the credibility of multiple candidate answers from an LLM, mitigating the model’s overconfidence in incorrect answers.

A survey of LLM-based agents for game playing.

Chain-of-thought prompting previously came in two flavors. Zero-shot CoT simply appends “Let’s think step by step” to the question, while Manual CoT (Few-shot CoT) supplies exemplars consisting of a question and its reasoning chain. The latter performs better, but only if the chains are well written — and writing them is manual labor. This paper proposes Auto-CoT, which generates few-shot CoT exemplars automatically.

Chain of thought (CoT) has already proven, in practice, to substantially improve the reasoning ability of large language models. Yet no work so far has explained the relationship between the length of a chain of thought and reasoning ability. This paper takes that question as its core, runs a systematic set of experiments on CoT, and arrives at a number of interesting and counterintuitive conclusions.

This paper proposes an LLM-based, graph-guided reasoning approach for multi-step reasoning problems. It makes two main contributions: the reasoning approach itself, and an in-context learning method for knowledge triplet extraction that allows variable definitions.

The ReAct paradigm: interleaving reasoning and acting in large language models.