Alex Murphy
@alexandermurphy.bsky.social
Student at the University of Edinburgh.
ML Theory, Adversarial ML, Optimization, Efficient ML
ML Theory, Adversarial ML, Optimization, Efficient ML
- Faithful decoding + ReAct consistently boosts performance on answering and recall across datasets and models
- They enhance ReAct's ability to follow the crucial Thought-Action-Observation format
- We achieved a significant Answer F1 improvement of up to 13.1 (using the DoLa decoder on HotpotQA)
- They enhance ReAct's ability to follow the crucial Thought-Action-Observation format
- We achieved a significant Answer F1 improvement of up to 13.1 (using the DoLa decoder on HotpotQA)
April 1, 2025 at 1:59 PM
- Faithful decoding + ReAct consistently boosts performance on answering and recall across datasets and models
- They enhance ReAct's ability to follow the crucial Thought-Action-Observation format
- We achieved a significant Answer F1 improvement of up to 13.1 (using the DoLa decoder on HotpotQA)
- They enhance ReAct's ability to follow the crucial Thought-Action-Observation format
- We achieved a significant Answer F1 improvement of up to 13.1 (using the DoLa decoder on HotpotQA)