V-SEAM combines concept-level visual semantic editing with attention head modulation to identify positive and negative contributors across object, attribute, and relationship levels, then uses this to improve VLM performance on VQA benchmarks.
arXiv preprint arXiv:2310.05916 , year=
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
representative citing papers
AnchorDiff performs training-free concept grounding in multi-modal diffusion transformers by anchor selection followed by graph propagation on attention-derived graphs, reducing concept leakage on a new multi-concept dataset.
Natural language descriptions generated via a closed-loop pipeline with digital twins capture the selectivity of most neurons in macaque V1 and V4, with synthesized images driving 96% of V4 neurons into the top or bottom 5% of natural-image response distributions.
Sparse feature circuits are introduced as interpretable causal subnetworks in language models, supporting unsupervised discovery of thousands of circuits and a method called SHIFT to improve classifier generalization by ablating irrelevant features.
AnimeAdapter is a modular adapter for Stable Diffusion that enables appearance-consistent anime character generation from a single reference image using semantic-selective local attention and pose-aware conditioning, plus a new Danbooru-derived dataset.
CLIP-SVD performs parameter-efficient adaptation of CLIP by fine-tuning singular values from SVD of weight matrices, reporting SOTA few-shot accuracy on 21 datasets plus a language-based interpretability analysis.
citing papers explorer
-
V-SEAM: Visual Semantic Editing and Attention Modulating for Causal Interpretability of Vision-Language Models
V-SEAM combines concept-level visual semantic editing with attention head modulation to identify positive and negative contributors across object, attribute, and relationship levels, then uses this to improve VLM performance on VQA benchmarks.
-
AnchorDiff: Training-Free Concept Grounding for MM-DiTs via Anchor-Based Graph Propagation
AnchorDiff performs training-free concept grounding in multi-modal diffusion transformers by anchor selection followed by graph propagation on attention-derived graphs, reducing concept leakage on a new multi-concept dataset.
-
Letting the neural code speak: Automated characterization of monkey visual neurons through human language
Natural language descriptions generated via a closed-loop pipeline with digital twins capture the selectivity of most neurons in macaque V1 and V4, with synthesized images driving 96% of V4 neurons into the top or bottom 5% of natural-image response distributions.
-
Sparse Feature Circuits: Discovering and Editing Interpretable Causal Graphs in Language Models
Sparse feature circuits are introduced as interpretable causal subnetworks in language models, supporting unsupervised discovery of thousands of circuits and a method called SHIFT to improve classifier generalization by ablating irrelevant features.
-
AnimeAdapter: A Modular Adapter for Appearance-Consistent Anime Character Generation
AnimeAdapter is a modular adapter for Stable Diffusion that enables appearance-consistent anime character generation from a single reference image using semantic-selective local attention and pose-aware conditioning, plus a new Danbooru-derived dataset.
-
CLIP-SVD: Efficient and Interpretable Vision-Language Adaptation via Singular Values
CLIP-SVD performs parameter-efficient adaptation of CLIP by fine-tuning singular values from SVD of weight matrices, reporting SOTA few-shot accuracy on 21 datasets plus a language-based interpretability analysis.