Hou Pong (Ken) Chan

Assistant Professor at the University of Macau @UM_1981 | Prev: @AlibabaGroup @uiuc_nlp @CUHKofficial

[Xueqing Wu @xueqing_w]

Human programmers post lessons on Stack Overflow, GitHub Gists, and blogs all the time - coding agents should do the same; knowledge discovered and then burried is a loss to the entire community.

Wrote a skill to encourage coding agents to publish lessons to GitHub Gist: https://t.co/1WyJPqldRT and will be using it across all my personal projects. People can also try it via: gh skill install shirley-wu/skills sharing-lessons --scope user --agent claude-code (or replace with your coding agent)

Happy for people to use it, share comments, and help it evolve.

[Hou Pong (Ken) Chan @kenchanhp]

@LegoX_team Very impressive work 👏

[LegoX @LegoX_team]

Thrilled to announce Lego-RL🚀, a harness-native reinforcement learning framework for coding agents like Claude Code, OpenCode and OpenHands-SDK. It successfully boosts Qwen-3.5-35B-A3B from 64% to 70% on SWE-Bench verified.

Everything is open-sourced: https://t.co/8yulA5J5pA

[Hou Pong (Ken) Chan @kenchanhp]

@shuyhere Congrats 👏

[Hou Pong (Ken) Chan @kenchanhp]

@Chenyang_Lyu Congrats Chenyang 👏

[Zhenhailong Wang @zhenhailongW]

Excited to share that I’ll join Westlake University as a tenure-track AP in AI from summer 2027! Welcome to join the lab as a "co-founder"! I am looking for: 2–3 PhD students (Fall 2027), RAs/interns. Check out the recruiting page for more details https://t.co/rMYwM2FyRJ :)

[Longyue Wang @wangly0229]

🏆 Best Resource Paper Award at #ACL2026 @aclmeeting

🚀Thrilled to share that our deep research benchmark paper "HSCodeComp: A Realistic and Expert-level Agent Benchmark for Hierarchical Rule Application" reveived a Best Resource Paper Award #ACL2026!

🎯 Hoping this sparks more focus on reliable AI Agents in the real world.

https://t.co/ToFSPPaYb3

[Denghui Zhang @denghui_zhang]

🕵️ Copyright Detective, #ACL2026 today at 12:45 PM PT. (virtual presentation 3)

We introduce an interactive forensic system for evidence discovery of copyrighted content leakage, not just static classification.

Live demo👇, welcome to try it!

[Yi R. (May) Fung @ECCV✈️ @May_F1_]

@JiayuJeff Great work from @JiayuLeff and the team, come check out our work on 𝙖𝙜𝙚𝙣𝙩 𝙘𝙤𝙨𝙩-𝙚𝙛𝙛𝙚𝙘𝙩𝙞𝙫𝙚𝙣𝙚𝙨𝙨 live at ACL'26! 🚀🚀

We also have follow-up work in the series worth mentioning as well, on "Adaptive Planning in LLM Agents under World & User Constraints". ✨✨

[Manling Li @ManlingLi_]

Welcome!

[Manling Li @ManlingLi_]

Tomorrow! Knowledgeable Foundation Models Workshop @ ACL 2026:

Zoom at https://t.co/Srbiv0DuXW 

9:30am PT - 5pm PT

[Min-Yen Kan @knmnyn]

Pls R/T to help! 🚨 Announcing a special issue on ethics in NLP (Computational Linguistics). We invite theory & tech subs on bias, fairness, power, multilinguality, decolonisation, and environmental/social impacts. 📅 DDL: 27 Nov 2026 #NLP #AIEthics

[Min-Yen Kan @knmnyn]

📣 Pls R/T 📅 COLING 2027 — 32nd Int’l Conf. on Computational Linguistics — Macau, China, May 9–14, 2027. Call for original long & short papers across all CL/NLP. Submit via ARR (Oct 2026 cycle); ddl Oct 12, 2026. Info: https://t.co/josPftoC4T #COLING2027

[Yi R. (May) Fung @ECCV✈️ @May_F1_]

📢 GroundLM 2026 @ #EMNLP2026 (Budapest, Oct 29)!

We're calling for papers on 𝗟𝗟𝗠 𝗳𝗮𝗶𝘁𝗵𝗳𝘂𝗹 & 𝗲𝗳𝗳𝗶𝗰𝗶𝗲𝗻𝘁 𝗴𝗿𝗼𝘂𝗻𝗱𝗶𝗻𝗴

Deadlines: Jun 29 (Direct) / Aug 5 (ARR)
🏆Also interesting shared tasks due later in Aug

Check out more below https://t.co/89R5f3uLeT

[Sumit @_reachsumit]

Understanding the Behaviors of Environment-aware Information Retrieval

Analyzes how LLMs learn to adapt query formulation to different retrievers via RL, showing that optimal query styles are retriever-specific.

📝 https://t.co/D3EznzAhBN
👨🏽‍💻 https://t.co/e9C7GPrsHM

[AACL 2026 @aaclmeeting]

The submission deadline for AACL-IJCNLP student research workshop is nearly a week away! #nlproc #aaclijcnlp

[Yichuan Wang @YichuanM]

The web was never meant to be flattened into text.

Yet most web RAG systems start by parsing HTML --- a complex and lossy process.

🔥 Introducing PixelRAG: the first RAG system that retrieves and reads 30M+ web pages as pixels.

Instead of extracting text, PixelRAG retrieves screenshots and lets a VLM read them directly.

PixelRAG not only preserves visual information, but also outperforms text-based RAG on text-only QA benchmarks by +18.1%.

Why?
(1) HTML-to-text conversion often discards layout, structure, tables, and other useful signals.
(2) We continued pretraining a VLM on web page screenshots and turned it into a surprisingly strong visual retriever.
(3) Recent VLMs are remarkably good at understanding web pages, often with better accuracy and token efficiency than text-only pipelines.

Takeaway: HTML parsing may be one of the biggest self-inflicted bottlenecks in web RAG.

Demo below 👇

Code: https://t.co/ssDF0nnVwZ
Paper: https://t.co/OIpQ26Vb8H
Playground: https://t.co/UdzM7GQmu3

[AACL 2026 @aaclmeeting]

AACL-IJCNLP 2026 Student Research Workshop (SRW) Pre-Submission Mentorship is now open! Details here: https://t.co/QbmjU6DRhJ

Pre-Submission Mentorship Deadline: June 8, 2026
Direct Submission Deadline: July 26, 2026

https://t.co/laGw3ZyMzb

#AACL2026 #NLProc

[Hou Pong (Ken) Chan @kenchanhp]

@LuWang__ @AmazonScience Congrats 🎉

[Lu Wang @LuWang__]

Thanks to @AmazonScience for supporting our work on measuring and monitoring scheming in AI agents. Looking forward to advancing tools for safer and more trustworthy AI systems!