CoEvoT: Co-Evolving Chain-of-Thought Prompting for Graph-LLM Reasoning
Quick Answer
CoEvoT introduces a novel co-evolving Chain-of-Thought prompting framework for graph-LLM reasoning, enhancing model adaptability under distribution shifts.
Quick Take
By integrating text-to-graph token rewriting with graph-to-text reasoning, it allows for dynamic evidence refinement, outperforming existing methods across eight datasets.
Key Points
- CoEvoT enables step-wise, state-aware evidence refinement in graph- reasoning.
- The framework couples text-to-graph token rewriting with graph-to-text reasoning guidance.
- Extensive experiments show CoEvoT outperforms state-of-the-art methods on eight datasets.
- Existing CoT-based methods are limited by fixed graph tokens during reasoning.
- CoEvoT enhances label-efficient predictions in graph learning tasks.
Paper Resources
📖 Reader Mode
~2 min readAbstract:Graph learning under distribution shift presents a persistent challenge, where models adapt to new graphs with limited or even no supervision. Recent graph--LLM approaches move toward label-efficient prediction by linearizing graphs into prompts and using large language models (LLMs) as predictors, and can adopt Chain-of-Thought (CoT) prompting to exploit LLM's multi-step reasoning capability. However, existing CoT-based graph--LLM methods generate intermediate thoughts while conditioning on fixed graph tokens, limiting step-wise refinement of structural cues. In this paper, we propose CoEvoT, a simple yet effective co-evolving CoT prompting framework for graph--LLM reasoning. CoEvoT couples text-to-graph token rewriting and graph-to-text reasoning guidance in a closed loop: each intermediate textual thought is used to update the graph token evidence state via a lightweight condition network, and the updated tokens are fed back into the next-step instruction to guide subsequent LLM reasoning. This enables step-wise, state-aware evidence refinement, rather than reasoning over a fixed graph snapshot. Extensive experiments on eight datasets demonstrate that CoEvoT consistently outperforms state-of-the-art baselines.
| Comments: | Under review |
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2607.14114 [cs.CL] |
| (or arXiv:2607.14114v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2607.14114 arXiv-issued DOI via DataCite |
Submission history
From: Xingtong Yu [view email]
[v1]
Fri, 8 May 2026 08:31:18 UTC (3,531 KB)
— Originally published at arxiv.org
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CL
See more →TriAgent: Divergence-Aware Committees for Cost-Efficient Financial Sentiment Analysis
TriAgent introduces a cost-efficient multi-agent system for financial sentiment analysis, combining VADER, FinBERT, and Qwen2.5. It achieves an F1 score of ~0.87 with significant savings of $9.3M/year at a 10M-user scale compared to GPT-4o-mini, while also detecting hallucinations with an AUC of 0.90.