AI Paper Atlas

Research map / Large language models / Reasoning and chain of thought

Structured Reasoning for LLMs: A Unified Framework for Efficiency and Explainability

Yubo Dong, Hehe Fan, Linchao Zhu, Yi Yang

ICLR 2026 · Large language models › Reasoning and chain of thought · approach: Search and self-correction in reasoning

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes Structured Reasoning, a framework inspired by cognitive science and neurosymbolic AI, aimed at improving LLM complex reasoning (logical deduction, planning) beyond token-level probability relationships, with claimed gains in efficiency and explainability.

Problem

LLMs struggle with complex reasoning such as logical deduction and planning because they rely primarily on token-level probability relationships.

Approach

Introduces Structured Reasoning, a unified framework that moves reasoning beyond token-level probabilities, drawing on cognitive science and neurosymbolic AI to improve both efficiency and explainability. Details are not available from the abstract alone.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add MaxFlow reward to GRPODeepSeek-R1-Distill-Qwen-1.5B · 9-benchmark average, 8K lengthavg pass@1: 43.28 -> 48.89Improved
add MaxFlow reward to GRPODeepSeek-R1-Distill-Qwen-7B · 9-benchmark average, 8K lengthavg pass@1: 58.33 -> 61.17Improved
add LCS reward to GRPODeepSeek-R1-Distill-Qwen-7B · 9-benchmark average, 8K lengthavg pass@1: 58.33 -> 57.65Hurt
add LCS reward to GRPO at short lengthsDeepSeek-R1-Distill-Qwen-7B · 9-benchmark average, 1K lengthavg pass@1: 29.19 -> 35.06Improved
use MaxFlow rewardDeepSeek-R1-Distill-Qwen-1.5B · temperature robustness, MATH500accuracy std: ±0.79 -> ±0.53Improved
use attention-based Max-Flow step importance instead of perplexity (ablation)step filtering · Error filtering efficiency, 9 stepsEFE: 68.07 -> 76.67Improved
use gold-order structured steps instead of shuffled (ablation)attention matrices · Entailment Trees task 1alignment: 28.48% -> 71.27%Improved
add structured guidance prompting without trainingDeepSeek-R1 671B · MATH500avg tokens: 2945 -> 1577Improved
add structured guidance prompting without trainingDeepSeek-R1 671B · AMC23avg tokens: 1716 -> 2053Hurt
combine Max-Flow and LCS rewards 1:1 (ablation)GRPO on 1.5B · Large Avg.pass@1: 58.1 (Max-Flow) vs 57.20 (joint)No clear effect
add MaxFlow reward to GRPO at 4K evaluation lengthDeepSeek-R1-Distill-Qwen-1.5B · 9-benchmark average, 4K lengthavg pass@1: 39.28 -> 41.68Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy