REVIEW 13 cited by
CodeI/O: Condensing Reasoning Patterns via Code Input-Output Prediction
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
CodeI/O: Condensing Reasoning Patterns via Code Input-Output Prediction
read the original abstract
Reasoning is a fundamental capability of Large Language Models. While prior research predominantly focuses on enhancing narrow skills like math or code generation, improving performance on many other reasoning tasks remains challenging due to sparse and fragmented training data. To address this issue, we propose CodeI/O, a novel approach that systematically condenses diverse reasoning patterns inherently embedded in contextually-grounded codes, through transforming the original code into a code input-output prediction format. By training models to predict inputs/outputs given code and test cases entirely in natural language as Chain-of-Thought (CoT) rationales, we expose them to universal reasoning primitives -- like logic flow planning, state-space searching, decision tree traversal, and modular decomposition -- while decoupling structured reasoning from code-specific syntax and preserving procedural rigor. Experimental results demonstrate CodeI/O leads to consistent improvements across symbolic, scientific, logic, math & numerical, and commonsense reasoning tasks. By matching the existing ground-truth outputs or re-executing the code with predicted inputs, we can verify each prediction and further enhance the CoTs through multi-turn revision, resulting in CodeI/O++ and achieving higher performance. Our data and models are available at https://github.com/hkust-nlp/CodeIO.
Forward citations
Cited by 13 Pith papers
-
Learning, Fast and Slow: Towards LLMs That Adapt Continually
Fast-Slow Training uses context optimization as fast weights alongside parameter updates as slow weights to achieve up to 3x better sample efficiency, higher performance, and less catastrophic forgetting than standard...
-
StepCodeReasoner: Aligning Code Reasoning with Stepwise Execution Traces via Reinforcement Learning
StepCodeReasoner aligns code reasoning with verifiable stepwise execution traces via print anchors and bi-level GRPO reinforcement learning, reaching SOTA results on CRUXEval (91.1%) and LiveCodeBench (86.5%) for a 7B model.
-
Think Anywhere in Code Generation
Think-Anywhere lets LLMs invoke on-demand reasoning at any token during code generation via cold-start imitation followed by outcome-based RL, reaching state-of-the-art results on LeetCode, LiveCodeBench, HumanEval, and MBPP.
-
Enhancing the Code Reasoning Capabilities of LLMs via Consistency-based Reinforcement Learning
CodeThinker improves LLM code reasoning via consistency-based RL with stepwise training data, dynamic beam sampling, and consistency rewards, reaching SOTA on benchmarks with 4.3% gains on Qwen2.5-Coder-7B.
-
Learning, Fast and Slow: Towards LLMs That Adapt Continually
Fast-Slow Training combines slow parameter updates with fast context optimization to achieve up to 3x better sample efficiency, higher performance, less forgetting, and preserved plasticity in continual LLM learning.
-
CodePivot: Bootstrapping Multilingual Transpilation in LLMs via Reinforcement Learning without Parallel Corpora
CodePivot uses Python as a pivot language plus an Aggressive-Partial-Functional RL reward to train a 7B model that outperforms much larger LLMs on multilingual code transpilation without parallel corpora.
-
Can Large Language Models Generalize Procedures Across Representations?
Post-training on graph or code versions of a planning task does not transfer to natural-language versions, but a symbolic-then-natural-language RL curriculum achieves strong transfer.
-
Generating Verifiable Chain of Thoughts from Exection-Traces
A pipeline produces 54,000 execution-trace-verified bi-directional Chain-of-Thought rationales for code, and fine-tuning on them yields gains up to 26.6 points on LiveCodeBench-Exec and similar benchmarks.
-
Rethinking Reasoning Quality in Large Language Models through Enhanced Chain-of-Thought via RL
DRER rewards CoT trajectories that increase the model's likelihood of the correct answer, plus a length penalty, and the new LogicTree benchmark reportedly lifts a 7B model's average accuracy from 0.13 to 0.60.
-
Dream-Coder 7B: An Open Diffusion Language Model for Code
An open 7B diffusion language model for code, adapted from Qwen2.5-Coder, reaches 21.4% pass@1 on LiveCodeBench and shows adaptive, non-left-to-right generation patterns.
-
Rethinking Burst Buffer Optimization: Enabling Layout Heterogeneity via Hybrid Analysis and LLM Guidance
Proteus combines static code analysis, runtime signals, and LLM guidance to select burst-buffer data layouts, claiming 91.3% decision accuracy and up to 3.24x/2.9x speedups on write- and metadata-intensive HPC workloads.
-
The Periodic Table of LLM Reasoning: A Structured Survey of Reasoning Paradigms, Methods, and Failure Modes
A literature survey that introduces a taxonomy for LLM reasoning paradigms, analyzes methodological trends, and synthesizes failure modes from over 300 papers.
-
Think in Games: Learning to Reason in Games via Reinforcement Learning with Large Language Models
A reinforcement-learning pipeline for predicting macro-actions in Honor of Kings improves action prediction accuracy, but the method is imitation of human replay labels, not the claimed environmental interaction.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.