At Upper Level Room 5AB of the conference venue!
San Diego, CA
Joined July 2025
- Tweets40
- Following0
- Followers203
- Likes29
Pinned Tweet
We’re excited to announce the first workshop on CogInterp: Interpreting Cognition in Deep Learning Models @ NeurIPS 2025! 📣
How can we interpret the algorithms and representations underlying complex behavior in deep learning models?
🌐 coginterp.github.io/neurips2…
1/
CogInterp Workshop @ NeurIPS 2025 retweeted
Can LLMs evolve human-like semantic categories?
CDS-affiliated @NogaZaslavsky and PhD student Nathaniel Imel show that, via simulated cultural transmission, LLMs reorganize color categories toward efficient compression.
🔗arxiv.org/abs/2509.08093
CogInterp Workshop @ NeurIPS 2025 retweeted
this slide is solid gold
Our last Stanford guest lecture - @EkdeepL on what counts as an explanation & a neuro-inspired "model systems approach" to interp
Plus, how in-context learning and many-shot jailbreaking are explained by LLM representations changing in-context (as a case study for that approach)
00:33 - What counts as an explanation?
04:47 - Levels of analysis & standard interpretability approaches
18:19 - The "model systems" approach to interp
[Case study on in-context learning]
23:36 - How LLM representations change in-context
44:10 - Modeling ICL with rational analysis
1:10:54 - Conclusion & questions
Thanks again to @SuryaGanguli for having us in his class!
CogInterp Workshop @ NeurIPS 2025 retweeted
Our last Stanford guest lecture - @EkdeepL on what counts as an explanation & a neuro-inspired "model systems approach" to interp
Plus, how in-context learning and many-shot jailbreaking are explained by LLM representations changing in-context (as a case study for that approach)
00:33 - What counts as an explanation?
04:47 - Levels of analysis & standard interpretability approaches
18:19 - The "model systems" approach to interp
[Case study on in-context learning]
23:36 - How LLM representations change in-context
44:10 - Modeling ICL with rational analysis
1:10:54 - Conclusion & questions
Thanks again to @SuryaGanguli for having us in his class!
CogInterp Workshop @ NeurIPS 2025 retweeted
Safety-oriented interpretability researchers should be focused on AI systems, not individual model artifacts. A snippet from the NeurIPS CogInterp workshop panel on Sunday:
CogInterp Workshop @ NeurIPS 2025 retweeted
Honored and thrilled that our work received the @CogInterp best paper award! 💫
📄 Extended paper: arxiv.org/pdf/2509.08093
🧵 Highlights: x.com/NogaZaslavsky/status/1…
@NeurIPSConf #NeurIPS2025
Our Best Paper Award goes to Nathaniel Imel and Noga Zaslavsky @NogaZaslavsky for their excellent paper “Culturally transmitted color categories in LLMs reflect a learning bias toward efficient compression”!
Our Best Paper Award goes to Nathaniel Imel and Noga Zaslavsky @NogaZaslavsky for their excellent paper “Culturally transmitted color categories in LLMs reflect a learning bias toward efficient compression”!
CogInterp Workshop @ NeurIPS 2025 retweeted
At the @CogInterp workshop at NeurIPS.
coginterp.github.io/neurips2…
This slide explains MechIntrep vs CongIntrep:
We are about to start our panel discussion, join us for some hot takes about what cognitive interpretability should be about.
Our final speaker @sydneymlevine makes a radical proposal: building computational models of human moral judgements to use as an AI system for making moral judgements.
Jay McClelland, opens with a question, "Do LMs have thoughts?" Are LMs stochastic parrots or is there some understanding?
Visualizing how LLMs handle object-property binding, he argues that even with scale, transformers might not be forming the kind of 'integrated representations' that human cognition relies on.
In our fourth spotlight talk, neural network legend Paul Smolensky uses symbolic programs such as production systems to understand how neural networks process symbols
For our third spotlight talk, Sonia Murthy @soniakmurthy uses probabilistic cognitive models to understand value trade-offs in LLMs that enable pragmatic reasoning about politeness in speech acts
Erin Grant @ermgrant discusses dissociations between function and representation, and asks whether representational alignment is enough for understanding deep neural networks
CogInterp Workshop @ NeurIPS 2025 retweeted
Excited to be presenting our work on using cognitive models to interpret pluralistic values in LLMs once again as a spotlight talk 🌟 at the NeurIPS CogInterp workshop!
Come by upper level room 5AB today and check out the paper here: arxiv.org/abs/2506.20666
The spotlight talks will cover all aspects of interpreting cognition in deep learning models: from behavior to algorithms to representations! Also check out the list of poster presentations at coginterp.github.io/neurips2…
(3/3)
