What do certified guarantees look like in the age of large language models and long reasoning chains? Look for us at EMNLP to find out!
04.11.2025 23:05 — 👍 1 🔁 0 💬 0 📌 0
Sum-of-Parts Models: Faithful Attributions for Groups of Features
Overcoming fundamental barriers in feature attribution methods with grouped attributions
If you're at ICML, in about 15 minutes, Weiqiu & I will be at our poster on sum-of-parts models: for faithful attributions and cosmology discovery. Stop by to say hi!
East Exhibition Hall A-B #E-1208
Thu 17 Jul 11 a.m. - 1:30 p.m. PDT
debugml.github.io/sum-of-parts/
#ICML @youweiqiu.bsky.social
17.07.2025 17:45 — 👍 1 🔁 1 💬 0 📌 0
LLM ignoring instructions? Make it listen with InstABoost.
✅ Simple: Steer your model in 5 lines of code
✅ Effective: Outperforms latent steering & prompt-only methods
✅ Grounded: Based on our mechanistic theory on rule-following (LogicBreaks)
Blog: debugml.github.io/instaboost
10.07.2025 18:46 — 👍 1 🔁 1 💬 0 📌 0
🧠 Foundation models are reshaping reasoning. Do we still need specialized neuro-symbolic (NeSy) training, or can clever prompting now suffice?
Our new position paper argues the road to generalizable NeSy should be paved with foundation models.
🔗 arxiv.org/abs/2505.24874
(🧵1/9)
13.06.2025 20:30 — 👍 1 🔁 1 💬 1 📌 0
Brachio Lab: FIX
We've been doing a bunch of interpretability work with scientists (i.e. our recent FIX benchmark brachiolab.github.io/fix/)!
21.11.2024 18:08 — 👍 2 🔁 0 💬 1 📌 0
