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
753 results for “keyword:"fine tuning"” in ICLR 2027. 158 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]Eigenbases Without Eigenvalues: Exact Amplification Laws for Spectral Preconditioning on Rank-Deficient Gradient Flows▲6.1k2.00
[PDF]Learning While Forgetting: Sparse Adapter Consolidation for Memory-Bounded Continual Learning▼4.5k2.67
[PDF]Static Gradient Attribution Underperforms a Density-Matched Random Mask Within LoRA’s B-Matrix▼1.2k1.67
[PDF]Steering Vectors Can Control What Models Learn in Supervised Fine-Tuning: An Analysis of Behavior Offloading▼1.2k1.67
[PDF]Mitigating Deceptive Behaviour in Large Language Models through Alignment and Unlearning Interventions▼5.8k2.33
[PDF]Understanding Forgetting Dynamics in Full Fine-Tuning of LLMs: A Neural Tangent Kernel Perspective▲3952.67
[PDF]SCouT: Softly Coupled Fine-Tuning for Preserving Specialization while Improving Mergeability▼4.3k2.00
[PDF]Understanding and Mitigating Forgetting in LLM Full Fine-Tuning at the Parameter Level▼4.3k2.00
[PDF]When Coding Skills Bind to Their Harness: Understanding and Mitigating Cross-Harness Failures in Agentic SFT▼4.3k2.00
[PDF]Searching the Quantized Weight Space for Adaptation: Task-Aware Joint Quantization and Low-Rank Initialization▲3.0k2.33
[PDF]RAC-LoRA: Router-Aware Combinatorial LoRA for Adaptive Parameter-Efficient Fine-Tuning▼4.3k2.00
[PDF]FoRA: Forward-Only Residual Adaptation for Federated Fine-Tuning of Large Language Models▲3952.67
[PDF]Stabilizing Zeroth-Order Fine-Tuning of Mixture-of-Experts via One-Sided Estimation and Expert Replay▲5103.00
Withdrawn or desk rejected (158)
[PDF]When to Act, Not What to Act: Prospective Authorization for Tool-Using Language ModelsWithdrawn1.33
[PDF]Why Benign Post-Training Induces Safety Gate Collapse: Changes in Harmful Activation Geometry Relative to the Inherited Refusal GateWithdrawn1.67