MASPrism attributes failures in LLM multi-agent executions by extracting token-level negative log-likelihood and attention weights from a small model's prefill pass, then ranking candidates with a second prefill, achieving top accuracy on most benchmarks and 6.69x speedup over baselines.
Attention is not not explanation
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
years
2026 6verdicts
UNVERDICTED 6representative citing papers
Multimodal contrastive learning using multilinear products is fragile to single bad modalities, and a gated version improves top-1 retrieval accuracy on synthetic and real trimodal data.
ORBIT uses an intervention-consistent self-supervised objective in a transformer to infer asymmetric gene program influences from observational scRNA-seq data, recovering Alzheimer's vulnerability patterns and achieving 0.984 macro F1 cell-type classification from 220 pathway scores.
Both humans and LLMs trust content more when labeled human-authored than AI-generated, with LLMs showing denser attention to labels and higher uncertainty under AI labels, mirroring human heuristic patterns.
Rigorous interpretability can function as a principled form of model evaluation if its claims are falsifiable, reproducible, and predictive.
Matched learning-rate experiments show LoRA retains substantially higher zero-shot transfer (45% vs 11% on EuroSAT, 58% vs 9% on Pets) than Full FT in CLIP adaptation.
citing papers explorer
-
MASPrism: Lightweight Failure Attribution for Multi-Agent Systems Using Prefill-Stage Signals
MASPrism attributes failures in LLM multi-agent executions by extracting token-level negative log-likelihood and attention weights from a small model's prefill pass, then ranking candidates with a second prefill, achieving top accuracy on most benchmarks and 6.69x speedup over baselines.
-
Hidden in the Multiplicative Interaction: Uncovering Fragility in Multimodal Contrastive Learning
Multimodal contrastive learning using multilinear products is fragile to single bad modalities, and a gated version improves top-1 retrieval accuracy on synthetic and real trimodal data.
-
ORBIT: Learning Gene Program Co-Activation Structure for Cell-Type-Stratified Pathway Rewiring Analysis in Single-Cell Transcriptomics
ORBIT uses an intervention-consistent self-supervised objective in a transformer to infer asymmetric gene program influences from observational scRNA-seq data, recovering Alzheimer's vulnerability patterns and achieving 0.984 macro F1 cell-type classification from 220 pathway scores.
-
Label Effects: Shared Heuristic Reliance in Trust Assessment by Humans and LLM-as-a-Judge
Both humans and LLMs trust content more when labeled human-authored than AI-generated, with LLMs showing denser attention to labels and higher uncertainty under AI labels, mirroring human heuristic patterns.
-
Rigorous Interpretation Is a Form of Evaluation
Rigorous interpretability can function as a principled form of model evaluation if its claims are falsifiable, reproducible, and predictive.
-
Matched-Learning-Rate Analysis of Attention Drift and Transfer Retention in Fine-Tuned CLIP
Matched learning-rate experiments show LoRA retains substantially higher zero-shot transfer (45% vs 11% on EuroSAT, 58% vs 9% on Pets) than Full FT in CLIP adaptation.