🧑‍🔬 Junjie Wang

🏛️ Affiliations

🏛️Chinese Academy of SciencesBeijing · China2026 · 2 paperscurrent🔬Institute of SoftwareBeijing · China2026 · 2 paperscurrent🎓Tsinghua UniversityBeijing · China2026 · 2 paperscurrent🎓University of Chinese Academy of SciencesBeijing · China2026 · 2 paperscurrent

From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine.

🧭 Fields, topics and concepts

Of its works in our library, with the number of works on each.

📅 Works by year

2020s

📄 Papers

The most cited first (OpenAlex and our library).

1. Solving Math Word Problems via Cooperative Reasoning induced Language Models (2023 · 86 citations)
   Pre-trained language models often fail on math word problems because they generate text without step-by-step supervision.
2. Chain-of-Reasoning: Towards Unified Mathematical Reasoning in Large Language Models via a Multi-Paradigm Perspective (2025 · 23 citations)
   Current LLM math solvers are usually tuned for a single reasoning style. Some use natural-language chains of thought, others use executable code, and others use formal proofs. As a result, a model strong at arithmetic word problems often fails at theorem proving, and the reverse. The authors propose…
3. ProFit: Leveraging High-Value Signals in SFT via Probability-Guided Token Selection (2026 · 4 citations)
   Standard supervised fine-tuning (SFT) trains an LLM to reproduce one reference answer token by token. Valid alternative phrasings are therefore penalized, and the model tends to memorize surface wording. Training on several references per prompt could help, but the authors find it expensive and unstable. They instead…
4. DEFT: Demystifying VLN Failures via a Unified Dual-View Explainability Framework for LLM-based Agents (2026)
   LLM planners in Vision-and-Language Navigation (VLN) are hard to diagnose. Existing explanation tools either find important timesteps or highlight important visual features, but not both together. They also handle the run-to-run randomness of black-box LLM agents poorly, and textual Chain-of-Thought is often…
5. GroupRank: A Groupwise Paradigm for Effective and Efficient Passage Reranking with LLMs (2026)
   LLM rerankers face a trade-off between accuracy and speed. Pointwise scoring is fast but never compares documents with each other. Listwise ranking does compare them, but it exceeds context limits and relies on slow sequential sliding windows. GroupRank takes a middle path: it splits the candidate pool into small…
6. OntoGuard: Enforcing Action Admissibility for LLM Agents in Complex Interactive Environments (2026)
   LLM agents in interactive simulators often hallucinate. They attempt actions that are impossible in the current state or that break the required order of steps. Methods based on imitation or sparse-reward learning only fit the statistics of the data and never represent these rules explicitly. OntoGuard is a…
7. WebClipper: Efficient Evolution of Web Agents with Graph-based Trajectory Pruning (2026)
   Open-source Deep Research web agents such as Tongyi-DeepResearch and MiroThinker reach high accuracy on complex information-seeking tasks.

🤝 Co-authors here

12 of 32 co-authors with a page, the most papers together first.

• Yujiu Yang ×3
• Dan Yang ×2
• Duolin Sun ×2
• Jie Feng ×2
• Meixiu Long ×2
• Yihan Jiao ×2
• Yue Shen ×2
• Yuxiang Zhang ×2
• Zhehao Tan ×2
• Binbin Hu
• Dongdong Zhang
• Furu Wei

Venues

Where their works here were published, the most works first, with the editions of those that have a page.

• Annual Meeting of the Association for Computational Linguistics (ACL) ×4: ACL 2026 ×2, ACL 2025, ACL 2023
• Findings of the Association for Computational Linguistics ×3: Findings of the Association for Computational Linguistics 2026 ×3

🔗 Cited by

Papers of our library cite their works 90 times (of the 149,439 papers whose bibliographies we read); 28 citing papers with a page here →.

Updated 1 November 2026

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/authors/junjie-wang/