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
25 results for “keyword:"internal representations"” in ICLR 2027. 11 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]Inference-Time Search in Discrete Diffusion Language Models, Guided by Internal-State Signals▼1.2k1.67
[PDF]Readable Constraints, Unreliable Decisions: A Constraint-Instantiation Lens on Language Model Errors▼1351.33
[PDF]What Does a Hallucination Score Tell Us? Auditing the Evidence from Prediction to Mechanistic Claims▼4.5k2.67
[PDF]When the Latest State Does Not Stick: STATEFLOW for Reliable Decisions over Evolving Records▼4.3k2.00
[PDF]Mind the Spike: Mechanisms and Brittleness of Visual Massive Activations in Large Vision–Language Models▼4.5k2.67
[PDF]Know Thyself, Teach Thyself: Internal Information Flow for Selective Self-Distillation▲5.4k3.00
Withdrawn or desk rejected (11)
[PDF]HPSA: Hallucination Detection from a Single Hidden State, Without Training or Layer SearchWithdrawn1.50
[PDF]Vision-Language Models Can Report Which Image's Visual-Token Activations Were PerturbedWithdrawn1.67
[PDF]Who Gets a Token, and What Does It Carry? Unequal Name Support and Concept Access in Large Language ModelsWithdrawn1.33
[PDF]Concepts Persist as Directions: Visual-State Auditing of Unlearned Diffusion ModelsWithdrawn1.67