USC NLP

👋 we’d love to join!

22.11.2024 19:09 — 👍 2    🔁 0    💬 0    📌 0

Auditing Political Exposure Bias: Algorithmic Amplification on Twitter/X Approaching the 2024 U.S. Presidential Election Approximately 50% of tweets in X's user timelines are personalized recommendations from accounts they do not follow. This raises a critical question: what political content are users exposed to beyond...

A rent paper has been portrayed in the news as proof that Twitter/X new #recsys favors republicans.

In our latest work, collecting millions of datapoints via 120 sock puppets, we should the reality is much more complex: you would be surprised by our findings:
arxiv.org/abs/2411.01852

thread 1/n

18.11.2024 05:57 — 👍 87    🔁 37    💬 13    📌 9

And we're having a great time at #EMNLP2024, come talk to us!

14.11.2024 23:34 — 👍 15    🔁 2    💬 0    📌 0

✨ Excited to share our Chain-of-Questions paper #EMNLP2023: we develop a framework that trains *one T5 model* to robustly answer multistep questions by generating and answering sub-questions. Outperforms ChatGPT on DROP, HotpotQA and their contrast/adversarial sets.

10.10.2023 22:57 — 👍 3    🔁 1    💬 2    📌 1

Logits of API-Protected LLMs Leak Proprietary Information The commercialization of large language models (LLMs) has led to the common practice of high-level API-only access to proprietary models. In this work, we show that even with a conservative assumption...

Just landed in Philly for COLM where I’ll be presenting my work on extracting secrets from LLM APIs at the Wednesday afternoon poster sesh.

Please reach out if you wanna hang and talk about sneaky LLM API hacks, accountability, and the geometry of LLM representations!

arxiv.org/abs/2403.09539

07.10.2024 03:51 — 👍 2    🔁 2    💬 0    📌 0