NeurIPS 2026 reviews are up: reviews, scores, rebuttals and meta-reviews for 7,877 papers. See the highest rated
acceptodds: the papers of ICLR 2027
172 results for “keyword:"sparse autoencoders"” in ICLR 2027. 33 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]Reconstruction, Not Localization: A Mechanistic Account of Sparse-Feature OOD Detection in Vision Transformers▼1361.33
[PDF]MechLeash-FL: Mechanistic Leashing of Latent Features for Federated Learning under Feature Skew▲402.80
[PDF]SFT Leaves a Reproducible Signature in Activation Geometry: Calibrating Post-Training Interpretability Against Run-to-Run Noise▼1.2k1.67
[PDF]De-Absorption Reparameterization: Post-hoc Repair of Feature Absorption in Sparse Autoencoders.▲3.0k2.33
[PDF]Unstable Features, Reproducible Subspaces: Understanding Seed Dependence in Sparse Autoencoders▼5.8k2.33
[PDF]Iterative distillation of Matryoshka sparse autoencoders improves hierarchical feature organization in ESMFold2▲11.00
[PDF]Exploring Divergence Between Brain Responses and Language Model Representations Using Crosscoders▲9.2k2.67
[PDF]SHADOW ANCHORING: GROUNDING HUMAN- INTERPRETABLE CONCEPTS IN TRANSFORMERS VIA SPARSE AUTOENCODERS▼1361.33
[PDF]A Linear Name-Origin Prior in the Recall Heads: How Language Models Fabricate and Misrecall Facts About People▲5103.00
[PDF]Conditional Hypothesis Generation for LLM-Based Text Analysis with Researcher-Specified Covariates▲13k2.33
[PDF]Cross-Architecture Universal SAEs: A Shared Sparse Dictionary for Vision and Diffusion Transformers▼1361.33
[PDF]Beyond Token Scale: Chunk-Level Sparse Autoencoders for Reliable Semantic Feature Discovery▲3.0k2.33
[PDF]CircuitSteer: Geometrically Aligned Multi-Layer Steering via Sparse Autoencoder Circuits▼4.5k2.67
[PDF]MAPP: Cross-Lingual Transfer through Representation Alignment with Performance-Gap Probing▲3.0k2.33
[PDF]Decodable but Misrouted: Sparse Features Uncover a Readout Gap in Vision-Language Models for Harmful Meme Detection▼1.2k1.67
[PDF]SAE-STM: A Sparse Autoencoder-Augmented Structural Topic Model for Interpretable Pairwise Preference Prediction▼1.2k1.67
[PDF]Which Features Matter? Sparse Latent Intervention for Prediction-Preserving Explanations▲5.4k3.00
[PDF]How Optimality Structures Sparse Dictionaries: Theory for Interpreting SAE Representations▲5.4k3.00