most citedPredictable Compression Failures: Order Sensitivity and Information Budgeting for Evidence-Grounded Binary Adjudication

1 citations · 1 across the 3 of their papers we have counts for

collaborators
Showing stat.MLShow all

2 papers · 1 filter

stat.ML2026

Attention Deficits in Language Models: Causal Explanations for Procedural Hallucinations

Ahmed Karim, Fatima Sheaib, Zein Khamis +3

Large language models can follow complex procedures yet fail at a seemingly trivial final step: reporting a value they themselves computed moments earlier. We study this phenomenon…

stat.ML2025★ 1 cited

Predictable Compression Failures: Order Sensitivity and Information Budgeting for Evidence-Grounded Binary Adjudication

Leon Chlon, Ahmed Karim, Maggie Chlon +1

Transformers used for evidence-grounded binary adjudication (e.g., support/refute, yes/no, or verifier-backed pass/fail decisions) can be sensitive to the order in which exchangeab…