On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,657 papers · 148 categories

Trend · papers per month

count % of corpus

2 results for “eval-x”

Have We Learned to Explain?: How Interpretability Methods Can Learn to Encode Predictions in their Interpretationsstat.ML

New method learns to encode predictions within interpretations, improving evaluation.

problem Need for interpretable machine learning, but existing methods are slow or lack fidelity.

method Amortized explanation methods that learn a global selector model optimizing fidelity of interpretations.

result Predictions can be encoded within interpretations, detected by EVAL-X.

interpretability machine learning fidelity predictions eval-x

2021-03-02abs ↗pdf ↗

Explanations that reveal all through the definition of encodingcs.LG

New definition reveals encoding explanations that retain predictive power.

problem Challenges in evaluating and identifying encoding explanations.

method Developed a definition of encoding based on conditional dependence.

result Existing evaluation scores do not rank non-encoding explanations correctly, but STRIPE-X does.

encoding explanations conditional dependence feature attributions predictive power LLM explanations

2024-11-04abs ↗pdf ↗