Challenge: Existing research to improve CoT efficiency falls into three categories, each with distinct limitations.
Approach: They propose a training-free framework that addresses both dimensions of CoT reasoning by applying a progressive precision reduction strategy coupled with an entropy-based confidence mechanism for adaptive termination.
Outcome: Empirical results show that the proposed framework achieves 11.3 efficiency gain without compromising accuracy.

Similar Papers

Graph-Based Chain-of-Thought Pruning for Reducing Redundant Reflections in Reasoning LLMs (2026.findings-acl)

Copied to clipboard

Challenge: Extending CoT through RL can induce undesirable thinking patterns such as overthinking . prior work has focused on inefficient reflection, which manifests in two problematic patterns: Indiscriminate Reflection and Repetitive Reflectione .
Approach: They propose a graph-based approach to optimize CoT by pruning each linear CoT into a directed acyclic graph with explicit dependency edges.
Outcome: The proposed approach reduces the average reasoning tokens by 42% while maintaining or improving accuracy.
CoTJudger: A Graph-Driven Framework for Automatic Evaluation of Chain-of-Thought Efficiency and Redundancy in LRMs (2026.findings-acl)

Copied to clipboard

Challenge: Existing evaluations emphasize final accuracy or coarse token counts, and lack automated tools to separate essential logic from structural redundancy.
Approach: They propose a graph-driven framework that quantifies reasoning efficiency by converting free-form CoTs into directed dependency graphs and extracting the Shortest Effective Path needed to reach a correct solution.
Outcome: Evaluating 21 LRMs, the proposed framework quantifies reasoning efficiency by converting free-form CoTs into directed dependency graphs and extracting the Shortest Effective Path (SEP) needed to reach a correct solution.
SoftCoT: Soft Chain-of-Thought for Efficient Reasoning with LLMs (2025.acl-long)

Copied to clipboard

Challenge: Existing approaches to continuous-space reasoning focus on hard token decoding and suffer from catastrophic forgetting.
Approach: They propose a method that generates instance-specific soft thought tokens as the initial chain of thoughts and maps them into the LLM’s representation space via a trainable projection module.
Outcome: The proposed method improves LLM reasoning performance through supervised, parameter-efficient fine-tuning.
Over-Reasoning and Redundant Calculation of Large Language Models (2024.eacl-short)

Copied to clipboard

Challenge: Large language models (LLMs) can solve problems step-by-step, but it is unclear whether they know when to use CoT and whether they are always necessary.
Approach: They propose to use LLMs to generate redundant calculations and reasoning on a manually constructed math QA dataset, GSM8K-Zero.
Outcome: The proposed model generates redundant calculations and reasoning on a manually constructed math QA dataset, but it is unclear whether it is necessary to use CoT reasoning.
Correct, Concise and Complete: Multi-stage Training For Adaptive Reasoning (2026.findings-acl)

Copied to clipboard

Challenge: Large language models (LLMs) increase test-time computation, often in the form of chain-of-thought (CoT) however, reasoning traces can become unnecessarily long, increasing computation costs without improving accuracy and sometimes even degrading performance.
Approach: They propose a multi-stage efficient reasoning method that combines supervised fine-tuning with reinforcement learning using an adaptive length penalty.
Outcome: The proposed method reduces response length by an average of 28% for 8B models and 40% for 32B models while incurring only minor performance drops of 1.6 and 2.5 points, respectively.
Think Faster Than Words: Efficient LLM Chain-of-Thought Reasoning via Dynamic Shortcut Decoding (2026.acl-long)

Copied to clipboard

Challenge: Existing methods that prune or employ early stopping to reduce latency often compromise reasoning reliability.
Approach: They propose a shortcut decoding framework that integrates probes over internal hidden states with step-level entropy to detect convergence of reasoning during generation and adaptively selects between a fast-exit path and a stability-verified path to remove redundant steps while preserving answer correctness.
Outcome: The proposed framework reduces token usage by approximately 35% and maintains accuracy comparable to full CoT decoding.
ReEfBench: Quantifying the Reasoning Efficiency of LLMs (2026.acl-long)

Copied to clipboard

Challenge: Existing methods for Chain-of-Thought evaluations do not distinguish between genuine reasoning and mere verbosity.
Approach: They propose a framework for the non-intrusive, comprehensive process-centric evaluation of reasoning grounded in First-Order Logic.
Outcome: The proposed framework identifies four distinct behavioral prototypes and diagnoses the failure modes.
Chain-of-Thought Degrades Visual Spatial Reasoning Capabilities of Multimodal LLMs (2026.acl-short)

Copied to clipboard

Challenge: Existing multimodal reasoning models lack generalized spatial intelligence, a new study shows . a critical gap exists in the field of vision-centric reasoning, the authors argue .
Approach: They evaluate 16 multimodal reasoning models using Chain-of-Though (CoT) based thinking . they find that CoT prompting consistently degrades performance in visual spatial reasoning .
Outcome: The proposed model hallucinates visual details from textual priors even when the image is absent.
AdapTime: Enabling Adaptive Temporal Reasoning in Large Language Models (2026.findings-acl)

Copied to clipboard

Challenge: Existing methods for temporal reasoning are limited and apply a fixed pipeline to all questions.
Approach: They propose an adaptive temporal reasoning method that dynamically executes reasoning steps based on context and task requirements.
Outcome: Experiments on two temporal QA benchmarks show the proposed method works.
Reinforced Efficient Reasoning via Semantically Diverse Exploration (2026.acl-long)

Copied to clipboard

Challenge: Existing methods for reinforcement learning with verifiable rewards suffer from limited exploration diversity and inefficient reasoning.
Approach: They propose a method that rewards concise and correct reasoning while penalizing unnecessarily long reasoning chains.
Outcome: Extensive experiments on Qwen and Llama models validate the effectiveness and efficiency of ROSE.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations