Our paper has been accepted at CogSci 2026! 🚀
The research demonstrates that aligning specific persona traits with task requirements significantly optimizes model attention and efficiency.
Find it at:
https://t.co/PUyi2vrahQ
I tore apart the Claude Code sourcecode and built claude-code-playbook — an interactive guide to how it’s designed, how it works, and what’s actually worth studying.
Site: https://t.co/oJcsXZbD5E
Repo: https://t.co/WtTegzPzet
@NeurIPSConf If the issue has been recognised, an apology should be issued as soon as possible and the relevant policies amended. If restrictions are necessary, why have universities on the list, like HIT and HEU, not been barred from submitting papers?
NeurIPS is aware of the community's concerns regarding the list of sanctions. NeurIPS is an inclusive community focused on free scientific discourse. We deeply value the research that comes from everyone in our community.
The present concerns are not about science or academic freedom. They are about legal requirements that apply to the NeurIPS Foundation, which is responsible for complying with sanctions. We are actively consulting legal counsel to fully understand the legal constraints and we will update the NeurIPS community as soon as we have reliable guidance from our lawyers.
As a potential author and reviewer, I do not want to review low-quality papers caused by a broken process, and I certainly do not want to receive careless or malicious reviews. Please fix the rule before it causes deeper damage to the community.
@NeurIPSConf There are now widespread calls online not just to boycott submissions/reviews, but to accept review assignments and then fail to review properly or review carelessly. If you want NeurIPS 2026 to proceed fairly and smoothly, please restore equal participation now.
Ready to give AI a soul? 🤖🧠
Join Psy-Symbiosis, the 1st Psychology AI Skill-a-thon by KinaMind! Build empathetic AI agents using TRAE—no complex coding needed. 🧬
🗓️ Mar 21, 2026 | 📍 SMU Singapore
Grab your spot today! 👇 🔗 https://t.co/SAcTRxePgI
#machinepsychology#TRAE
Win up to S$500 this weekend in Singapore! Join KinaMind Psy-Symbiosis AI Hackathon on Mar 21 at SMU. Build empathetic AI agents using TRAE with zero coding required. Even OpenClaw the lobster needs skills, so lets build them together! Register now: https://t.co/SAcTRxePgI
New Article: "T-COL: Generating Counterfactual Explanations for General User Preferences on Variable Machine Learning Systems" by Wang, Wang, Wu, Feng, and Zhang https://t.co/OWvWts5e1E
To simulate realistic counseling cases, we use specialized LoRAs on Qwen2.5: Emotion Inferencer & Chief Chain Generator
With @GradientHQ's Parallax, we now serve these multiple LoRAs on a SINGLE base model instance. Huge memory savings & seamless distributed support! ⚡️
We won 1st Place at the TRAE Friends Hackathon with AnnaAgent! 🏆
Grateful to the hosts & the TRAE framework for helping us evolve our paper into a full app.
AI is a tool to empower human counselors, not replace them.
#CSAIA#traefriendsSingapore#AnnaAgent
@JefferyTatsuya After receiving some issues and PRs, I recognized that the community needs not just a reimplementation. So I converted this repo to a Context Engineering Toolkit. The new link: https://t.co/OAkQ3SshjH. Welcome anyone to contribute!
@JefferyTatsuya Thank you for your forward. If there are any errors or discrepancies in my project, you can report issues or submit Pull Requests directly. Also, welcome everyone who wants to contribute to reproducing the project of this paper or other high-quality papers.💯