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
73 results for “keyword:"compositional generalization"” in ICLR 2027. 24 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]Winning Without Composing, Losing While Generalizing: Separating Better Scores from Better Choices in Recommendation▼5.8k2.33
[PDF]Latent Assembly: Causal Discovery and Cross-Task Linking of Executable Operations in Language Models▲13k2.33
[PDF]Beyond One Missing Hyperfact: Witness-Guided Full-Inference Complex Query Answering over Hyper-Relational Knowledge Graphs▼4.3k2.00
[PDF]When More Data Is Not Enough: Identifying and Repairing Support Gaps in Mechanism-Resolved Forecasting▲9.2k2.67
[PDF]POSS: Learning from Partially Observed Object States for Compositional Zero-Shot Learning▼1.2k3.00
[PDF]Input Embeddings as a Compositional Pronunciation Interface for Frozen LLM-Based Chinese TTS▼4.3k2.00
[PDF]Beyond Activity Prediction: Evaluating Compositional Regulatory Generalization in Genomic Foundation Models▲5103.00
[PDF]Zero-Shot Compositional Generalization in Neural PDE Solvers via Conservative Material-Guided Routing▼5.8k2.33
[PDF]TEXT AS A COMPOSITIONAL INTERFACE: TOWARD DECOMPOSABLE LEARNING WITH RECON- STRUCTED POLICIES▼1.2k1.67
[PDF]RAVEL: Factorizing multi-region neural communication into reusable propagation routes and dynamic recruitment▼1.2k1.67
[PDF]ScaleComposer: Scale-Conditioned Compositional Learning for Source-Count Extrapolation in Physical Fields▼4.5k2.67
[PDF]Readable Factors, Unresolved Combinations: Support-Aware Correction for Warm-Start Recommendation▼4.3k2.00
[PDF]The Clocks of Composition: Built by the Signal, Lost to Weight Decay, Kept by Rehearsal▼1.2k1.67
[PDF]Familiar Operations, Unseen Structures: How Task Compositions Affect Task Generalization?▲3.0k2.33
[PDF]LLMs Can Follow Instructions, But Not Many at Once: Multiplicative Collapse in Compositional Instruction Following▼1.2k1.67
[PDF]Same States, Different Actuators: What Single-Actuator Data Must Cover for Compositional World Models▲6.1k2.00
Withdrawn or desk rejected (24)
[PDF]DACL: Predicting Responses to Pairs of Unseen Genes by Reusing Single-Perturbation SupervisionWithdrawn1.67
[PDF]Dissecting Reinforcement Learning: Mechanisms Behind Compositional Reasoning in LLMsWithdrawn1.67
[PDF]Morphosemantic Concept Systems: Learning Reusable Discrete Concepts for Compositional GeneralizationWithdrawn1.67
[PDF]Knowing the Parts Is Not Enough: Measuring Composition After Matched Precursor SuccessDesk rejected
[PDF]When Stronger Baselines Mislead: Admissibility Auditing for Representation-Level ClaimsDesk rejected