←
Runjin Chen15 posts
Runjin Chen avatar

Runjin Chen

@RunjinChen

Member of Technical Staff @AnthropicAI | PH.D. student @UTAustin @VITAGroupUT | Previously BS/MS @sjtu1896

◷ Joined August 2023
Runjin Chen avatar
Runjin Chen@RunjinChen·1 Aug 2025

New Anthropic Research: Persona Vectors We can: 1. Monitor how a model’s personality is changing during a conversation, or over training 2. Mitigate undesirable persona shifts during development or prevent during training. 3. Identify training data that leads to shift

New Anthropic research: Persona vectors. Language models sometimes go haywire and slip into weird and unsettling personas. Why? In a new paper, we find “persona vectors"—neural activity patterns controlling traits like evil, sycophancy, or hallucination.

Post attachment
Jack Lindsey avatar
⟳ Runjin Chen retweeted
Jack Lindsey@Jack_W_Lindsey·Apr 7

Before limited-releasing Claude Mythos Preview, we investigated its internal mechanisms with interpretability techniques. We found it exhibited notably sophisticated (and often unspoken) strategic thinking and situational awareness, at times in service of unwanted actions. (1/14)

Post attachment
Anthropic avatar
⟳ Runjin Chen retweeted
Anthropic@AnthropicAI·Apr 7

Introducing Project Glasswing: an urgent initiative to help secure the world’s most critical software. It’s powered by our newest frontier model, Claude Mythos Preview, which can find software vulnerabilities better than all but the most skilled humans. anthropic.com/glasswing

Victor.Kai Wang avatar
⟳ Runjin Chen retweeted
Victor.Kai Wang@VictorKaiWang1·Mar 5

very happy to release this parameter generation work. from P-diff (2024), RPG (2025), DnD (2025) to HY-WU, parameter generation becomes more and more practical. thx @TencentHunyuan and happy to work with @oahzxl @mmbronstein @ZiqiaoWang63428 @cindy_x_wu @YangYou1991 @VITAGroupUT

One static model does not fit all😭 We just dropped our latest work: Functional Neural Memory. Instead of static models, we generate custom "parameters" for every single input. ✅Prompt your model anytime ✅Instant personalization ✅Better instruction following ✅Flexible & dynamic memory (w/o memory bank✌️) (🧵1/6)

0:09
Junyuan "Jason" Hong avatar
⟳ Runjin Chen retweeted
Junyuan "Jason" Hong@hjy836·19 Oct 2025

🧠 Conclusion: Data curation is cognitive hygiene for AI. 🩺 Regular data “health checks” are essential for keeping models reliable, safe, and aligned. 💡 The striking parallels between AI and human cognitive decline may even offer new insights into human brain health. 👩‍🔬 Work by: Shuo Xing*, Junyuan Hong*, Yifan Wang, Runjin Chen, Zhenyu Zhang, Ananth Grama, Zhengzhong Tu, and Zhangyang Wang 🌐 Website (code, data, models): llm-brain-rot.github.io 📄 ArXiv: arxiv.org/abs/2510.13928

Ethan Perez avatar
⟳ Runjin Chen retweeted
Ethan Perez@EthanJPerez·4 Sep 2025

We’re hiring someone to run the Anthropic Fellows Program! Our research collaborations have led to some of our best safety research and hires. We’re looking for an exceptional ops generalist, TPM, or research/eng manager to help us significantly scale and improve our collabs 🧵

Emmanuel Ameisen avatar
⟳ Runjin Chen retweeted
Emmanuel Ameisen@mlpowered·1 Aug 2025

In which the gang (@RunjinChen, @andyarditi, @Jack_W_Lindsey ): - identifies vectors for bad personas (evil, sycophancy, hallucinations, etc) - shows that if you inject the bad vectors in training, the model learns to not do the bad thing!! aka vaccines but for LLMs

New Anthropic research: Persona vectors. Language models sometimes go haywire and slip into weird and unsettling personas. Why? In a new paper, we find “persona vectors"—neural activity patterns controlling traits like evil, sycophancy, or hallucination.

Post attachment
Anthropic avatar
⟳ Runjin Chen retweeted
Anthropic@AnthropicAI·1 Aug 2025

New Anthropic research: Persona vectors. Language models sometimes go haywire and slip into weird and unsettling personas. Why? In a new paper, we find “persona vectors"—neural activity patterns controlling traits like evil, sycophancy, or hallucination.

Post attachment
Runjin Chen avatar
Runjin Chen@RunjinChen·20 Feb 2024

Our LLaGA excels in versatility, generalizability and interpretability, allowing it to perform consistently well across different datasets and tasks, extend its ability to unseen datasets or tasks, and provide explanations for graphs

Runjin Chen avatar
Runjin Chen@RunjinChen·20 Feb 2024

Key Feature: A versatile linear projector seamlessly bridges graph structures with the token space understood by Large Language Models (LLMs).