iiinova — where your digital self keeps evolving

Anmol Goel

@anmolgoel.bsky.social

156 followers 970 following 11 posts

NLP ∩ Privacy PhD @ TU Darmstadt x UCopenhagen goel.ai

Posts Replies Media

Reposted by Anmol Goel

Parameter Lab @parameterlab.bsky.social · 03/02/2026

‼️New paper from Parameter Lab! ⛓️‍💥 We identify privacy collapse, a silent failure mode of LLMs: LLMs fine-tuned on seemingly benign data can lose their ability to respect contextual privacy norms. Done by @anmolgoel.bsky.social during his internship! Check-out 👇

051

Reposted by Anmol Goel

Martin Gubri @mgubri.bsky.social · 03/02/2026

New paper out!🎉 One of our most surprising findings: fine-tuning an LLM on debugging code has unexpected side-effects on contextual privacy. The model learns from printing variables that internal state are ok to share, then generalises this to social situations🤯 A🧵below👇

072

Anmol Goel @anmolgoel.bsky.social · 03/02/2026

🚨 Fine-tuning your model to be more helpful or empathetic might be making it less private, without you noticing. In our latest work, we show that benign fine-tuning can silently break contextual privacy in language models while safety & general capabilities appear intact. ⬇️

140

Reposted by Anmol Goel

UKP Lab @ukplab.bsky.social · 27/01/2025

#ICLR »Differentially Private Steering for Large Language Model Alignment« by @anmolgoel.bsky.social, Yaxi Hu, Iryna Gurevych (@igurevych.bsky.social) & Amartya Sanyal (@amartyasanyal.bsky.social) (2/🧵)

163

Reposted by Anmol Goel

Christoph Molnar @christophmolnar.bsky.social · 22/11/2024

Just realized BlueSky allows sharing valuable stuff cause it doesn't punish links. 🤩 Let's start with "What are embeddings" by @vickiboykis.com The book is a great summary of embeddings, from history to modern approaches. The best part: it's free. Link: vickiboykis.com/what_are_emb...

Over the past decade, embeddings — numerical representations of
machine learning features used as input to deep learning models — have
become a foundational data structure in industrial machine learning
systems. TF-IDF, PCA, and one-hot encoding have always been key tools
in machine learning systems as ways to compress and make sense of
large amounts of textual data. However, traditional approaches were
limited in the amount of context they could reason about with increasing
amounts of data. As the volume, velocity, and variety of data captured
by modern applications has exploded, creating approaches specifically
tailored to scale has become increasingly important.
Google’s Word2Vec paper made an important step in moving from
simple statistical representations to semantic meaning of words. The
subsequent rise of the Transformer architecture and transfer learning, as
well as the latest surge in generative methods has enabled the growth
of embeddings as a foundational machine learning data structure. This
survey paper aims to provide a deep dive into what embeddings are,
their history, and usage patterns in industry.

22647101

Reposted by Anmol Goel

Chen Cecilia Liu @ccliu.bsky.social · 21/11/2024

Sorry that I’m missing a lot of people. If you’re working on making NLP models more culturally aware, please DM me to be added. go.bsky.app/tRMpng

2123

Reposted by Anmol Goel

Amartya Sanyal @amartyasanyal.bsky.social · 21/11/2024

I made a starter pack for european researchers interested in some aspects of learning theory. The list is clearly inexhaustive. So please enter your suggestions in comments. go.bsky.app/5o5uVnr

1122