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
4,880 results for “keyword:"arge language models"” in ICLR 2027. 989 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]Learning to Manage Context: Training Agents with Meta-Tools Beyond a Single Context Window▲3.0k2.33
[PDF]TEB: A Bandwidth-Constrained Evidence Allocation Framework for Enhancing Multimodal Sentiment Analysis and Emotion Recognition in Large Language Models▼1.2k1.67
[PDF]Content-Matched Is Not Lexically Matched: Text and Untrained-Network Baselines for Evaluation-Awareness Probing▼4.5k2.67
[PDF]Extending Controllability from Composition and Topology to Domain Identity in Language-Model Protein Generation▲5103.00
[PDF]LAB-Sparse: Local Accumulative Boost for Adaptive Semantic-Block Retention in KV Cache▼4.3k2.00
[PDF]When LLMs Bid Rationally, and When They Don't: A Traceable Capability Split in Sealed-Bid Auctions▼1361.33
[PDF]OUTPUT WITHOUT WARRANT: SPECTRAL DIMENSION READOUTS CANNOT CERTIFY THEIR OWN HYPOTHESES▲6.1k2.00
[PDF]Counterfactual Error Scrambling: Measuring Error-Message Use in Tool-Using Language Agents▼4.3k2.00
[PDF]RANA: Accelerating Multi-Turn Agentic Reinforcement Learning with Speculative Decoding▼4.3k2.00
[PDF]Early Context Starvation in Conversational LLMs: Causal Evidence from Attention Interventions▼1.2k1.67
[PDF]BEAR-Med: From a Clinically Grounded Benchmark to Expert-Aligned Reasoning in Medical Image Forensics▲5.4k3.00
[PDF]Probing Under Distribution Shift: Learning Generalizable Distributed Representations of LLM Activations▼5.9k2.33
[PDF]Enhancing Robustness of Small-Scale Models for Text-to-SQL via Semantic Enhancement and Noise-Aware Methods▼1.2k1.67
[PDF]One Token Flips the Verdict: Hypergraph Propagation and State Readout Dissociate in Selective Belief Revision▼5.9k2.50
[PDF]Large Language Model for Combinatorial Optimization by Leveraging Policy Adaptation and Heuristics▼71.25
[PDF]Disentangling Verbalized Evaluation Awareness and Refusal through Neuron-Level Intervention▼4.3k2.00
[PDF]Many Caches, Same Reasoning: Rethinking Importance Selection for Reasoning-Time KV Compression▼1.2k1.67
Withdrawn or desk rejected (989)
[PDF]TACE: Two-Stage Advantage and Credit Estimation for Multimodal Reinforcement Learning with Verifiable RewardsWithdrawn
[PDF]AnchorCredit: Output-Anchored Credit Assignment for Structured Generation with Multiple Verifiable RewardsDesk rejected2.67
[PDF]Where Does the Video Go? Probing Persistent Temporal Memory after Visual KV Eviction in Video LLMsWithdrawn2.00
[PDF]When Objectives Agree But Rules Do not: The Hidden Cost of Exposing Rule Errors In Optimization ModelsWithdrawn1.00
[PDF]Pursuing the Pareto Frontier for Interactive Intervention with LLM-based Reward DesignWithdrawn1.67