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
208 results for “keyword:"architectures"” in ICLR 2027. 44 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]Learning to Copy, Learning to Depend: Language Models Offload In-Context Copying to a Pointer Head▼1.3k1.75
[PDF]Early Context Starvation in Conversational LLMs: Causal Evidence from Attention Interventions▼1.2k1.67
[PDF]Would an Untrained Encoder Do as Well? The Missing Baseline in Graph Self-Supervised Learning▼4.3k2.00
[PDF]Motion-JEPA: Physically Structured Residual Predictive Representation Learning for Autonomous Driving▼1.2k1.67
[PDF]Branch-JEPA: Learning Dynamics-Aware State Representations through Dual-Branch Patch Routing▼1361.33
[PDF]Towards All-Atom Protein World Models: Learning and Diagnosing Predictive Latent Representations▼4.3k2.00
[PDF]Which State Should Prediction Read? A Mechanistic Analysis of State-Prediction Separation▲5.4k3.00
[PDF]Representational Collapse and Spectral Recovery in Koopman-Structured Time-Series JEPAs▼1.2k1.67
[PDF]Temporal Action Structure Controls Representation Identifiability in Predictive World Models▲3.0k2.33
[PDF]Vestrum: Improving Agent Harnesses by Adapting Their Verification, Structure and Memory▲13k2.33
[PDF]LeDreamer: Bridging the Understanding of Representation Collapse Between JEPA and Sample-Efficient World Models▼5.8k2.33
[PDF]IsoAlign: Isoform-Aware DNA-Protein Alignment via Joint-Embedding Predictive Architectures▼4.3k2.00
[PDF]A Faithful Attribution Reports Its Head, Not Its Model: The Limits of Completeness in Additively Decomposed Representations▼1.2k1.67
[PDF]The COUNT: Counting N-grams Unlocks Scaling on Host Memory and Training-Free Adaptation▼4.5k2.67
[PDF]HEPA: A Self-Supervised Horizon-Conditioned Event Predictive Architecture for Time Series▲3.0k2.50
Withdrawn or desk rejected (44)
[PDF]InnerJEPA: Masked Visual Latent Prediction for Video Understanding and Generative TransferWithdrawn2.00
[PDF]PL-JEPA: Joint Predictive and Generative Learning for Structure-Based Drug DiscoveryDesk rejected2.00
[PDF]The Conjunctive Fraction of Embodied Benchmarks: Composition Cannot Be Measured Where the Write Path Has Already Destroyed ItWithdrawn1.33
[PDF]OpenArch: A Unified PyTorch Framework for Implementing Modern Open-Source LLM ArchitecturesDesk rejected
[PDF]Loop LeWorldModel: Reliable Recursion for Depth Reuse and Temporal Rollout in Latent World ModelsWithdrawn1.67