Mind the Gaps: From Failure Attribution to Closed-Form Repair of Code Language Models
cs.SE, cs.CL, cs.LG
Submitted: 2026-10-04
Updated: 2026-10-04
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- Eliciting Latent Predictions from Transformers with the Tuned Lens
- Evaluating Large Language Models Trained on Code
- Vocabulary-Defined Semantics: Latent Space Clustering for Improving In-Context Learning
- Beyond Neural Incompatibility: Cross-Scale Knowledge Transfer in Language Models through Latent Semantic Alignment
- Rethinking Weight Tying: Pseudo-Inverse Tying for LM Stable Training and Updates
- SemRF: A Semantic Reference Frame for Residual-Stream Dynamics in Language Models
- MiniCPM: Unveiling the Potential of Small Language Models with Scalable Training Strategies
- RustEvo^2: An Evolving Benchmark for API Evolution in LLM-based Rust Code Generation
- CodeUpdateArena: Benchmarking Knowledge Editing on API Updates
- In-context Learning and Induction Heads
- VersiCode: Towards Version-controllable Code Generation
Related papers
- Falsification-Based Verification of LLM-Generated Optimization Models: Sound Test Batteries and Their Detection Limits
- GitSkills: A Dataset of Agent Skills on GitHub
- SABER: Benchmarking Operational Safety of LLM Coding Agents in Stateful Project Workspaces
- PackMonitor: Enabling Zero Package Hallucinations Through Decoding-Time Monitoring
- IntentCoding: Amplifying User Intent in Code Generation
- Incentives and Outcomes in Bug Bounties