7 papers · 1 filter
LESSER: Post-Training Data Selection with Output-Layer Gradients
Lyuxin David Zhang, Eric Wong, Surbhi Goel +1
The choice of post-training data for large language models substantially affects downstream performance. Gradient-based data selection is a popular approach that ranks training dat…
Missingness Bias Calibration in Feature Attribution Explanations
Shailesh Sridhar, Anton Xue, Eric Wong
Popular explanation methods often produce unreliable feature importance scores due to missingness bias, a systematic distortion that arises when models are probed with ablated, out…
AnCoder: Anchored Code Generation via Discrete Diffusion Models
Anton Xue, Litu Rout, Constantine Caramanis +1
Diffusion language models offer a compelling alternative to autoregressive code generation, enabling global planning and iterative refinement of complex program logic. However, exi…
Probabilistic Soundness Guarantees in LLM Reasoning Chains
Weiqiu You, Anton Xue, Shreya Havaldar +4
In reasoning chains generated by large language models (LLMs), initial errors often propagate and undermine the reliability of the final conclusion. Current LLM-based error detecti…
Probabilistic Stability Guarantees for Feature Attributions
Helen Jin, Anton Xue, Weiqiu You +2
Stability guarantees have emerged as a principled way to evaluate feature attributions, but existing certification methods rely on heavily smoothed classifiers and often produce co…
AR-Pro: Counterfactual Explanations for Anomaly Repair with Formal Properties
Xiayan Ji, Anton Xue, Eric Wong +2
Anomaly detection is widely used for identifying critical errors and suspicious behaviors, but current methods lack interpretability. We leverage common properties of existing meth…
The FIX Benchmark: Extracting Features Interpretable to eXperts
Helen Jin, Shreya Havaldar, Chaehyeon Kim +10
Feature-based methods are commonly used to explain model predictions, but these methods often implicitly assume that interpretable features are readily available. However, this is…