Nishant Subramani @ ACL
@nsubramani23.bsky.social
PhD student @CMU LTI - working on model #interpretability, student researcher @google; prev predoc @ai2; intern @MSFT
nishantsubramani.github.io
nishantsubramani.github.io
At @colmweb.org all week 🥯🍁! Presenting 3 mechinterp + actionable interp papers at @interplay-workshop.bsky.social
1. BERTology in the Modern World w/ @bearseascape.bsky.social
2. MICE for CATs
3. LLM Microscope w/ Jiarui Liu, Jivitesh Jain, @monadiab77.bsky.social
Reach out to chat! #COLM2025
1. BERTology in the Modern World w/ @bearseascape.bsky.social
2. MICE for CATs
3. LLM Microscope w/ Jiarui Liu, Jivitesh Jain, @monadiab77.bsky.social
Reach out to chat! #COLM2025
October 6, 2025 at 10:08 PM
At @colmweb.org all week 🥯🍁! Presenting 3 mechinterp + actionable interp papers at @interplay-workshop.bsky.social
1. BERTology in the Modern World w/ @bearseascape.bsky.social
2. MICE for CATs
3. LLM Microscope w/ Jiarui Liu, Jivitesh Jain, @monadiab77.bsky.social
Reach out to chat! #COLM2025
1. BERTology in the Modern World w/ @bearseascape.bsky.social
2. MICE for CATs
3. LLM Microscope w/ Jiarui Liu, Jivitesh Jain, @monadiab77.bsky.social
Reach out to chat! #COLM2025
MICE 🐭:
🎯 - significantly beats baselines on expected tool-calling utility, especially in high risk scenarios
✅ - matches expected calibration error of baselines
✅ - is sample efficient
✅ - generalizes zeroshot to unseen tools
5/🧵
🎯 - significantly beats baselines on expected tool-calling utility, especially in high risk scenarios
✅ - matches expected calibration error of baselines
✅ - is sample efficient
✅ - generalizes zeroshot to unseen tools
5/🧵
April 29, 2025 at 1:41 PM
MICE 🐭:
🎯 - significantly beats baselines on expected tool-calling utility, especially in high risk scenarios
✅ - matches expected calibration error of baselines
✅ - is sample efficient
✅ - generalizes zeroshot to unseen tools
5/🧵
🎯 - significantly beats baselines on expected tool-calling utility, especially in high risk scenarios
✅ - matches expected calibration error of baselines
✅ - is sample efficient
✅ - generalizes zeroshot to unseen tools
5/🧵
We propose 🐭 MICE to better assess confidence when calling tools:
1️⃣ decode from each intermediate layer of an LM
2️⃣ compute similarity scores between each layer’s generation and the final output.
3️⃣ train a probabilistic classifier on these features
3/🧵
1️⃣ decode from each intermediate layer of an LM
2️⃣ compute similarity scores between each layer’s generation and the final output.
3️⃣ train a probabilistic classifier on these features
3/🧵
April 29, 2025 at 1:41 PM
We propose 🐭 MICE to better assess confidence when calling tools:
1️⃣ decode from each intermediate layer of an LM
2️⃣ compute similarity scores between each layer’s generation and the final output.
3️⃣ train a probabilistic classifier on these features
3/🧵
1️⃣ decode from each intermediate layer of an LM
2️⃣ compute similarity scores between each layer’s generation and the final output.
3️⃣ train a probabilistic classifier on these features
3/🧵
🚀 Excited to share a new interp+agents paper: 🐭🐱 MICE for CATs: Model-Internal Confidence Estimation for Calibrating Agents with Tools appearing at #NAACL2025
This was work done @msftresearch.bsky.social last summer with Jason Eisner, Justin Svegliato, Ben Van Durme, Yu Su, and Sam Thomson
1/🧵
This was work done @msftresearch.bsky.social last summer with Jason Eisner, Justin Svegliato, Ben Van Durme, Yu Su, and Sam Thomson
1/🧵
April 29, 2025 at 1:41 PM
🚀 Excited to share a new interp+agents paper: 🐭🐱 MICE for CATs: Model-Internal Confidence Estimation for Calibrating Agents with Tools appearing at #NAACL2025
This was work done @msftresearch.bsky.social last summer with Jason Eisner, Justin Svegliato, Ben Van Durme, Yu Su, and Sam Thomson
1/🧵
This was work done @msftresearch.bsky.social last summer with Jason Eisner, Justin Svegliato, Ben Van Durme, Yu Su, and Sam Thomson
1/🧵