Bing He

Publications [Google Scholar]

(* denotes equal contribution; # denotes mentored interns/students.)
Code, data, and downloadable datasets are linked below wherever available.

Peer-Reviewed Papers

  1. (EMNLP 2026) Minhua Lin#, Juncheng Wu#, Zijun Wang#, Zhan Shi, Yisi Sang, Bing He, et al., “Harness Updating Is Not Harness Benefit: Disentangling Evolution Capabilities in Self-Evolving LLM Agents”, code and data, paper, The Conference on Empirical Methods in Natural Language Processing 2026, Main Conference RSIAgent HarnessEvaluation & Benchmark

  2. (EMNLP 2026) Tianxin Wei#, Zhan Shi, Minhua Lin#, Bing He, et al., “Evo-Harness: Context-to-Harness Skill Compilation for Self-Evolving Agents”, code and data, paper, The Conference on Empirical Methods in Natural Language Processing 2026, Main Conference RSIAgent HarnessAgents & Tool Use

  3. (ICLR 2026) Pengfei He#, Zhenwei Dai, Bing He, et al., “TRAJECT-Bench: A Trajectory-Aware Benchmark for Evaluating Agentic Tool Use”, code and data, paper, The International Conference on Learning Representations 2026 Agents & Tool UseEvaluation & Benchmark

  4. (ACL 2026) Yingqian Cui#, Zhenwei Dai, Pengfei He, Bing He, et al., “A Reward-Guided Dual-Phase Framework for Adaptive Inference-Time Reasoning”, paper, The 64th Annual Meeting of the Association for Computational Linguistics 2026, Main Findings Reasoning RLLLM Reasoning

  5. (NeurIPS 2026 Workshop) Zewen Liu#, Zhan Shi, Yisi Sang, Bing He, et al., “Adaptive Auto-Harness: Sustained Self-Improvement on Open-Ended Task Streams”, code and data, paper, The Workshop on Continual Learning for Enterprise AI Agents (CLEA) at the Conference on Neural Information Processing Systems 2026, Workshop Poster RSIAgent Harness

  6. (CAIS 2026 Workshop) Minhua Lin#, Hanqing Lu, Zhan Shi, Bing He, et al., “Position: Agentic Evolution is the Path to Evolving LLMs”, code and data, paper, The First Workshop on Agent Skills at the ACM Conference on AI and Agentic Systems 2026, Workshop Poster RSIAgent Harness

  7. (TKDD 2025) Bing He, Yibo Hu, Yeon-Chang Lee, et al., “A Survey on the Role of Crowds in Combating Online Misinformation: Annotators, Evaluators, and Creators”, paper collections, survey, ACM Transactions on Knowledge Discovery from Data, January 2025 Trust & SafetyNLP

  8. (WWW 2024) Bing He, Sreyashi Nag, Limeng Cui, et al., “Hierarchical Query Classification in E-commerce Search”, paper, The ACM Web Conference 2024 (Acceptance Rate: 21.3%) NLPLLM Application

  9. (WebSci 2024) Bing He, Yingchen Ma, Mustaque Ahamad, Srijan Kumar, “Corrective or Backfire: Characterizing and Predicting User Response to Social Correction”, code and data, downloadable dataset, The ACM Web Science Conference 2024 Trust & SafetyLLM JudgeEvaluation & Benchmark

  10. (WWW 2023) Bing He, Mustaque Ahamad, Srijan Kumar, “Reinforcement Learning-based Counter-Misinformation Response Generation: A Case Study of COVID-19 Vaccine Misinformation”, code and data, paper, downloadable dataset, The ACM Web Conference 2023 (Acceptance Rate: 365/1900=19.2%), Best Paper Nomination LLM Post-TrainingReasoning RLData Curation, Pipeline & RL EnvironmentNLPTrust & Safety

  11. (WebSci 2023) Yingchen Ma#, Bing He, Nathan Subrahmanian, Srijan Kumar, “Characterizing and Predicting Social Correction on Twitter”, code and data, paper, downloadable dataset, The ACM Web Science Conference 2023 Trust & SafetyLLM JudgeEvaluation & Benchmark

  12. (KDD 2021) Bing He, Mustaque Ahamad, Srijan Kumar, “PETGEN: Personalized Text Generation Attack on Deep Sequence Embedding-based Classification Models”, code and data, paper, ACM SIGKDD Conference on Knowledge Discovery and Data Mining 2021 (Acceptance Rate: 238/1541=15.4%) NLPTrust & Safety

  13. (ASONAM 2021) Bing He, Caleb Ziems, Sandeep Soni, Naren Ramakrishnan, Diyi Yang, Srijan Kumar, “Racism is a virus: Anti-asian Hate and Counterspeech in Social Media during the Covid-19 Crisis”, code and data, paper, downloadable dataset (tweets), (geolocations), IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining 2021 NLPTrust & Safety

  1. (IEEE BigData 2020) Bing He*, Nicholas Micallef*, Srijan Kumar, Mustaque Ahamad, Nasir Memon, “The Role of the Crowd in Countering Misinformation: A Case Study of the COVID-19 Infodemic”, code and data, paper, downloadable dataset (annotated tweets), (raw tweets), IEEE International Conference on Big Data 2020 NLPTrust & Safety

  2. (ECML-PKDD 2018) Bing He, Xiaolin Chen, Dian Zhang, Siyuan Liu, Dawei Han, Lionel M Ni, “PBE: Driver Behavior Assessment Beyond Trajectory Profiling”, paper, The European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases 2018 Data MiningMachine Learning

  3. (IEEE BigData 2018) Bing He, Dian Zhang, Siyuan Liu, Hao Liu, Dawei Han, Lionel M Ni, “Profiling driver behavior for personalized insurance pricing and maximal profit”, paper, IEEE International Conference on Big Data 2018 Data MiningMachine Learning

  4. (ICCP 2014) Junyu Lai, Bing He, Huawei Chen, Wenhong Tian, Kaiyu Qin, “A Channel Popularity Oriented Transmission Scheme in Vehicular IPTV Networks”, paper, IEEE International Conference on Communication Problem-solving 2014 Networking

  5. (IC-NIDC 2014) Junyu Lai, Bing He, Guolin Sun, Weiwei Fang, Kaiyu Qin, “Improve Channel Availability in Vehicular IPTV Systems by a Hybrid Transmission Scheme”, paper, IEEE International Conference on Network Infrastructure and Digital Content 2014 Networking

  6. (CNIS 2014) Junyu Lai, Bing He, Guolin Sun, Gun Li, Kaiyu Qin, “Enabling Software-Defined Networking Technologies in Carrier Networks”, International Conference on Computer Network and Information Science 2014 Networking

Preprints

  1. (arXiv 2026) Minhua Lin#, Zijun Wang#, Zewen Liu#, Zhaotian Weng, Bing He, et al., “Language Models Learn to Improve Language Models”, paper, arXiv preprint arXiv:2610.23344, 2026 RSIAutoresearchLLM Post-Training

  2. (arXiv 2026) George Ma, Yuhao Zhang, Bing He, et al., “Verifier-Grounded Agentic RL for Provably Correct Code”, paper, arXiv preprint arXiv:2610.19298, 2026 Agentic RLCoding Agent

  3. (arXiv 2026) Guanghan Wang, Bing He, et al., “Revisiting the 16x Pretraining Compute Gap between Diffusion and Autoregressive Language Models”, project page, paper, arXiv preprint arXiv:2610.17590, 2026 Pretraining & Diffusion LMs

  4. (arXiv 2026) Zhisheng Qi, Haoyu Han, Jiaxi Li, Bing He, et al., “Co-adaptive Rule Routing and Constraining for Structured Knowledge Retrieval”, paper, arXiv preprint arXiv:2610.14465, 2026 Reasoning RLAgentic RL

  5. (arXiv 2026) Zijun Wang#, Zewen Liu#, Minhua Lin#, Zhaotian Weng, Zhan Shi, Bing He, et al., “Language Models as AI Research World Models”, paper, arXiv preprint arXiv:2610.12235, 2026 RSIAutoresearch

  6. (arXiv 2026) Yiqiao Jin, Yiyang Wang, Lucheng Fu, Bing He, et al., “Science Utopia? Closed-Loop LLM Simulation of Academic Research Ecosystems”, code, project page, paper, arXiv preprint arXiv:2610.01257, 2026 Agents & Tool Use

  7. (arXiv 2026) Rui Sun, Zhan Shi, Bing He, “TRACE: Training Reasoning Agents for Causal Exploration with Synthesized Rewards”, paper, arXiv preprint arXiv:2609.10315, 2026 Agentic RLReasoning RLData Curation, Pipeline & RL EnvironmentCoding Agent

  8. (arXiv 2026) Zhan Shi, Bing He, et al., “A-Evolve-Training: Autonomous Post-Training of a 30B Model”, paper, arXiv preprint arXiv:2606.20657, 2026 RSIAutoresearchLLM Post-TrainingData Curation, Pipeline & RL Environment

  9. (arXiv 2026) Yingqian Cui#, Zhenwei Dai, Bing He, et al., “How Do Latent Reasoning Methods Perform Under Weak and Strong Supervision?”, paper, arXiv preprint arXiv:2602.22441, 2026 LLM Latent Reasoning

  10. (arXiv 2025) Bing He, Mustaque Ahamad, Srijan Kumar, “ROBAD: Robust Adversary-aware Local-Global Attended Bad Actor Detection Sequential Model”, paper, arXiv preprint arXiv:2507.15067, 2025 Trust & SafetyLLM Post-Training

Thesis and Book Chapter

  1. (Ph.D. Thesis 2024) Bing He, “Leveraging AI to Combat Misinformation by Empowering Crowds and Evaluating Detectors”, Ph.D. Dissertation, Georgia Institute of Technology, 2024 Trust & SafetyNLPLLM Application

  2. (Book Chapter 2024) Srijan Kumar, Bing He, “Crowd-Powered Solutions to Identify and Correct Online Misinformation”, The BARONS and the MOB, 2024 Trust & Safety

Blogs

  1. (Blog 2026) Bing He, Rui Sun, Zhan Shi, et al., “How Good Are GLM5.2 and Kimi-K3: Create Your Own Benchmark and Revisit Your Training Recipe”, blog, Jun 2026 (updated Jul 2026 with Kimi-K3 results) Evaluation & BenchmarkAgents & Tool UseLLM Post-TrainingAgentic RL

  2. (Blog 2026) Bing He, “Designing Self-evolving Coding Agents for Terminal-Bench 2.0: A Systematic Exploration”, blog, Apr 2026 RSIAgent HarnessCoding AgentEvaluation & Benchmark

  3. (Blog 2026) Bing He, Rui Sun, Zhan Shi, Hanqing Lu, “From Llama-2 Replication to DeepSeek V3.2&R1: Revisiting DeepSeek’s 13 Research Papers for the Complete Technical Evolution of DeepSeek Base and Reasoning Models”, blog, Feb 2026 Pretraining & Diffusion LMsLLM Post-TrainingReasoning RLLLM Reasoning

  4. (Blog 2025) Bing He, Rui Sun, Zhan Shi, “Building and Aligning LLM using User Data and RL”, blog, Nov 2025 LLM Post-TrainingData Curation, Pipeline & RL EnvironmentLLM Application

  5. (Blog 2025) Zhan Shi, Rui Sun, Bing He, “Awesome Cursor Training Tutorial”, blog, Nov 2025 Agentic RLCoding AgentData Curation, Pipeline & RL Environment

  6. (Blog 2025) Zhan Shi, Rui Sun, Bing He, “The Hidden Language of AI: How Chat Templates Reveal the Evolution of LLMs”, blog, Oct 2025 LLM Post-TrainingAgents & Tool UseNLP