Skip to content
← Back to explorer

Draft-Thinking: Learning Efficient Reasoning in Long Chain-of-Thought LLMs

Jie Cao, Tianwei Lin, Zhenxuan Fan, Bo Yuan, Ziyuan Zhao, Rolan Yan, Wenqiao Zhang, Siliang Tang · Feb 28, 2026 · Citations: 0

Open arXiv Find Implementation RSS feed Shortlist (0)

Data freshness

Extraction: Fresh

Check recency before relying on this page for active eval decisions. Use stale pages as context and verify against current hub results.

Metadata refreshed

Feb 28, 2026, 9:57 AM

Recent

Extraction refreshed

Mar 8, 2026, 2:44 AM

Fresh

Extraction source

Runtime deterministic fallback

Confidence 0.30

Abstract

Long chain-of-thought~(CoT) has become a dominant paradigm for enhancing the reasoning capability of large reasoning models~(LRMs); however, the performance gains often come with a substantial increase in reasoning budget. Recent studies show that existing CoT paradigms tend to induce systematic overthinking, unnecessarily coupling reasoning capability with reasoning cost. Most prior approaches reduce token usage through post hoc techniques such as token compression, truncation, or length penalties, without explicitly addressing the core mechanisms of reasoning. We propose \textbf{Draft-Thinking}, which guides models to first learn a concise \textit{draft-style} reasoning structure that retains only the critical reasoning steps. Through a \textit{progressive curriculum learning}, the model stably internalizes this efficient reasoning pattern as its capability scales. Moreover, Draft-Thinking introduces adaptive prompting, which elevates reasoning depth to a flexible, model-selectable behavior. Extensive experiments demonstrate that Draft-Thinking substantially reduces reasoning budget while largely preserving reasoning performance; for example, on MATH500, it achieves an 82.6\% reduction in reasoning budget at the cost of only a 2.6\% performance drop.

Low-signal caution for protocol decisions

Use this page for context, then validate protocol choices against stronger HFEPX references before implementation decisions.

  • Extraction flags indicate low-signal or possible false-positive protocol mapping.
  • Extraction confidence is 0.30 (below strong-reference threshold).
  • No explicit evaluation mode was extracted from available metadata.
Human Eval Hub LLM-as-Judge Hub Pairwise Preference Hub

HFEPX Relevance Assessment

This paper is adjacent to HFEPX scope and is best used for background context, not as a primary protocol reference.

Best use

Background context only

Use if you need

A benchmark-and-metrics comparison anchor.

Main weakness

Extraction flags indicate low-signal or possible false-positive protocol mapping.

Trust level

Low

Eval-Fit Score

0/100 • Low

Treat as adjacent context, not a core eval-method reference.

Human Feedback Signal

Not explicit in abstract metadata

Evaluation Signal

Weak / implicit signal

HFEPX Fit

Adjacent candidate

Extraction confidence: Low

If you are doing eval pipeline work, start here:

Human Eval Hub LLM-as-Judge Hub Pairwise Preference Hub Tool-Use Eval Hub

Field Provenance & Confidence

Each key protocol field shows extraction state, confidence band, and data source so you can decide whether to trust it directly or validate from full text.

Human Feedback Types

missing

None explicit

Confidence: Low Source: Runtime deterministic fallback missing

No explicit feedback protocol extracted.

Evidence snippet: Long chain-of-thought~(CoT) has become a dominant paradigm for enhancing the reasoning capability of large reasoning models~(LRMs); however, the performance gains often come with a substantial increase in reasoning budget.

Evaluation Modes

missing

None explicit

Confidence: Low Source: Runtime deterministic fallback missing

Validate eval design from full paper text.

Evidence snippet: Long chain-of-thought~(CoT) has become a dominant paradigm for enhancing the reasoning capability of large reasoning models~(LRMs); however, the performance gains often come with a substantial increase in reasoning budget.

Quality Controls

missing

Not reported

Confidence: Low Source: Runtime deterministic fallback missing

No explicit QC controls found.

Evidence snippet: Long chain-of-thought~(CoT) has become a dominant paradigm for enhancing the reasoning capability of large reasoning models~(LRMs); however, the performance gains often come with a substantial increase in reasoning budget.

Benchmarks / Datasets

partial

MATH 500, DROP

Confidence: Low Source: Runtime deterministic fallback evidenced

Useful for quick benchmark comparison.

Evidence snippet: Extensive experiments demonstrate that Draft-Thinking substantially reduces reasoning budget while largely preserving reasoning performance; for example, on MATH500, it achieves an 82.6\% reduction in reasoning budget at the cost of only a 2.6\% performance drop.

Reported Metrics

partial

Cost

Confidence: Low Source: Runtime deterministic fallback evidenced

Useful for evaluation criteria comparison.

Evidence snippet: Recent studies show that existing CoT paradigms tend to induce systematic overthinking, unnecessarily coupling reasoning capability with reasoning cost.

Rater Population

missing

Unknown

Confidence: Low Source: Runtime deterministic fallback missing

Rater source not explicitly reported.

Evidence snippet: Long chain-of-thought~(CoT) has become a dominant paradigm for enhancing the reasoning capability of large reasoning models~(LRMs); however, the performance gains often come with a substantial increase in reasoning budget.

Human Data Lens

  • Uses human feedback: No
  • Feedback types: None
  • Rater population: Unknown
  • Unit of annotation: Unknown
  • Expertise required: Math
  • Extraction source: Runtime deterministic fallback

Evaluation Lens

  • Evaluation modes:
  • Agentic eval: None
  • Quality controls: Not reported
  • Confidence: 0.30
  • Flags: low_signal, possible_false_positive, runtime_fallback_extraction

Protocol And Measurement Signals

Benchmarks / Datasets

MATH-500DROP

Reported Metrics

cost

Research Brief

Deterministic synthesis

We propose Draft-Thinking, which guides models to first learn a concise draft-style reasoning structure that retains only the critical reasoning steps. HFEPX protocol signal is limited in abstract-level metadata, so treat it as adjacent context. Updated from current HFEPX corpus.

Generated Mar 8, 2026, 2:44 AM · Grounded in abstract + metadata only

Key Takeaways

  • We propose Draft-Thinking, which guides models to first learn a concise draft-style reasoning structure that retains only the critical reasoning steps.
  • Abstract shows limited direct human-feedback or evaluation-protocol detail; use as adjacent methodological context.

Researcher Actions

  • Treat this as method context, then pivot to protocol-specific HFEPX hubs.
  • Cross-check benchmark overlap: MATH-500, DROP.
  • Validate metric comparability (cost).

Caveats

  • Generated from title, abstract, and extracted metadata only; full-paper implementation details are not parsed.
  • Low-signal flag detected: protocol relevance may be indirect.

Research Summary

Contribution Summary

  • We propose Draft-Thinking, which guides models to first learn a concise draft-style reasoning structure that retains only the critical reasoning steps.

Why It Matters For Eval

  • Abstract shows limited direct human-feedback or evaluation-protocol detail; use as adjacent methodological context.

Researcher Checklist

  • Gap: Human feedback protocol is explicit

    No explicit human feedback protocol detected.

  • Gap: Evaluation mode is explicit

    No clear evaluation mode extracted.

  • Gap: Quality control reporting appears

    No calibration/adjudication/IAA control explicitly detected.

  • Pass: Benchmark or dataset anchors are present

    Detected: MATH-500, DROP

  • Pass: Metric reporting is present

    Detected: cost

Category-Adjacent Papers (Broader Context)

These papers are nearby in arXiv category and useful for broader context, but not necessarily protocol-matched to this paper.

Need human evaluators for your AI research? Scale annotation with expert AI Trainers.

Post a Job Get a Quote