allenai/ContextEval
Preview
•
Updated
•
63
•
3
Nils Feldhus PRO
nfel
AI & ML interests
Interpretability, Explainability, Natural Language Generation
Recent Activity
liked
a dataset
about 7 hours ago
allenai/ContextEval
liked
a dataset
15 days ago
TheBlueScrubs/TheBlueScrubs-v1
upvoted
a
paper
27 days ago
Enhancing Automated Interpretability with Output-Centric Feature
Descriptions
Organizations
nfel's activity
liked
a
dataset
about 7 hours ago
liked
a
dataset
15 days ago
upvoted
a
paper
27 days ago
liked
a
model
about 1 month ago
upvoted
a
paper
about 2 months ago
liked
a
Space
about 2 months ago
liked
a
model
3 months ago
liked
a
Space
6 months ago
authored
3
papers
9 months ago
Thermostat: A Large Collection of NLP Model Explanations and Analysis Tools
Paper
•
2108.13961
•
Published
Saliency Map Verbalization: Comparing Feature Importance Representations from Model-free and Instruction-based Methods
Paper
•
2210.07222
•
Published
InterroLang: Exploring NLP Models and Datasets through Dialogue-based Explanations
Paper
•
2310.05592
•
Published
upvoted
a
paper
11 months ago
liked
2
models
12 months ago
liked
a
Space
12 months ago
liked
a
Space
about 1 year ago
reacted to
gsarti's
post with 👍
about 1 year ago
Post
🔍 Today's pick in Interpretability & Analysis of LMs: SyntaxShap: Syntax-aware Explainability Method for Text Generation by
@kamara000
, R. Sevastjanova and M. El-Assady
Most model-agnostic post-hoc interpretability methods used nowadays in NLP were originally ported from tabular/CV domains with next to no adjustments to the intrinsic properties of textual inputs.
In this work, authors propose SyntaxSHAP, an adaptation of the Shapely value approach in which coalitions used to compute marginal contributions to importance scores are constrained by the syntax of the explained sentence. The resulting tree-based coalitions do not satisfy the efficiency assumption of Shapley values but preserves the symmetry, nullity and additivity axioms.
SyntaxSHAP is compared to other model-agnostic approaches on small (GPT-2 117M) and large (Mistral 7B) LMs, showing it produces explanations that are more faithful to model predictions and more semantically meaningful than other common methods, while also being more efficient than the base SHAP method.
📄 Paper: SyntaxShap: Syntax-aware Explainability Method for Text Generation (2402.09259)
💻 Code: https://github.com/k-amara/syntax-shap
🔍 All daily picks in LM interpretability: https://huggingface.co/collections/gsarti/daily-picks-in-interpretability-and-analysis-of-lms-65ae3339949c5675d25de2f9
Most model-agnostic post-hoc interpretability methods used nowadays in NLP were originally ported from tabular/CV domains with next to no adjustments to the intrinsic properties of textual inputs.
In this work, authors propose SyntaxSHAP, an adaptation of the Shapely value approach in which coalitions used to compute marginal contributions to importance scores are constrained by the syntax of the explained sentence. The resulting tree-based coalitions do not satisfy the efficiency assumption of Shapley values but preserves the symmetry, nullity and additivity axioms.
SyntaxSHAP is compared to other model-agnostic approaches on small (GPT-2 117M) and large (Mistral 7B) LMs, showing it produces explanations that are more faithful to model predictions and more semantically meaningful than other common methods, while also being more efficient than the base SHAP method.
📄 Paper: SyntaxShap: Syntax-aware Explainability Method for Text Generation (2402.09259)
💻 Code: https://github.com/k-amara/syntax-shap
🔍 All daily picks in LM interpretability: https://huggingface.co/collections/gsarti/daily-picks-in-interpretability-and-analysis-of-lms-65ae3339949c5675d25de2f9
liked
2
models
about 1 year ago
