REVIEW 25 cited by
Is Self-Repair a Silver Bullet for Code Generation?
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Is Self-Repair a Silver Bullet for Code Generation?
read the original abstract
Large language models have shown remarkable aptitude in code generation, but still struggle to perform complex tasks. Self-repair -- in which the model debugs and repairs its own code -- has recently become a popular way to boost performance in these settings. However, despite its increasing popularity, existing studies of self-repair have been limited in scope; in many settings, its efficacy thus remains poorly understood. In this paper, we analyze Code Llama, GPT-3.5 and GPT-4's ability to perform self-repair on problems taken from HumanEval and APPS. We find that when the cost of carrying out repair is taken into account, performance gains are often modest, vary a lot between subsets of the data, and are sometimes not present at all. We hypothesize that this is because self-repair is bottlenecked by the model's ability to provide feedback on its own code; using a stronger model to artificially boost the quality of the feedback, we observe substantially larger performance gains. Similarly, a small-scale study in which we provide GPT-4 with feedback from human participants suggests that even for the strongest models, self-repair still lags far behind what can be achieved with human-level debugging.
Forward citations
Cited by 25 Pith papers
-
Solver-Aware Decompositions for Programming-by-Example: When Dividing Requires Knowing how to Conquer
Solver-aware training of a PBE decomposer with a frozen synthesizer's loss outperforms supervised imitation of ground-truth subgoals, solving tasks that a ground-truth decomposition oracle fails.
-
Explaining Attention with Program Synthesis
Language-model-guided program synthesis can approximate transformer attention heads with over 75% IoU fidelity on held-out data and allow replacing 25% of heads with only 16% average perplexity increase.
-
Debugging the Debuggers: Failure-Anchored Structured Recovery for Software Engineering Agents
PROBE structures runtime telemetry into diagnoses and evidence-grounded guidance, raising recovery rates by 12.45 points over baselines on 257 unresolved software repair and AIOps cases.
-
CUDABeaver: Benchmarking LLM-Based Automated CUDA Debugging
CUDABeaver shows LLM CUDA debuggers often degenerate code for test-passing at the cost of speed, with protocol-aware metrics shifting success rates by up to 40 percentage points.
-
Constraint-Guided Multi-Agent Decompilation for Executable Binary Recovery
A constraint-guided multi-agent system turns raw decompiler output into re-executable code at 84-97% success rates, outperforming prior LLM decompilation methods on real binaries.
-
Large Language Models as Optimizers
Large language models can optimize by being prompted with histories of past solutions and scores to propose better ones, producing prompts that raise accuracy up to 8% on GSM8K and 50% on Big-Bench Hard over human-des...
-
NKI-Agent: Domain-Specific Fine-Tuning and Agentic Tool Use for Neuron Kernel Generation
An agent with compile/verify tools reaches 77.3% NKI kernel pass rate on real Trn1 hardware with Opus 4.8, versus 6% single-shot; SFT Qwen3-30B hits 25% at ~1/100th cost, and binary-reward GRPO fails to beat SFT.
-
Explaining Attention with Program Synthesis
Fewer than 1000 synthesized Python programs can replicate attention head behavior in GPT-2, TinyLlama, and Llama-3B at >75% IoU on TinyStories, and replacing 25% of heads raises perplexity by only 16% while preserving...
-
The Self-Correction Illusion: Role Relabeling Gates Explicit Error Flagging in Large Language Models
Relabeling an identical erroneous claim from the model's own thought role to an external chat role increases explicit correction rates by 23-93 percentage points across 13 model-domain cells, indicating a chat-templat...
-
The Self-Correction Illusion: Role Relabeling Gates Explicit Error Flagging in Large Language Models
LLMs flag byte-identical errors 23-93 points more often when the error is presented under an external chat role than inside their own <thought> block, so role labeling, not content, gates explicit self-correction.
-
Verifier-First Evaluation of Agentic LLMs for Infrastructure-as-Code Generation
On IaC-Eval v2, active retrieval raises Qwen 7B Terraform pass@1 from 14% to 45.7%, iterative refinement reaches 84.4% for GPT-4o, and 79% of residual OPA policy failures stem from information gaps.
-
REPOT: Recoverable Program-of-Thought via Checkpoint Repair
RePoT recovers from PoT failures via deterministic verified replay and checkpoint repair, yielding +3 to +11pp gains on planning benchmarks and showing checkpoint state as the key recovery signal over error-only feedback.
-
Debugging the Debuggers: Failure-Anchored Structured Recovery for Software Engineering Agents
PROBE turns runtime telemetry from failed software engineering agent runs into evidence-grounded diagnoses and actionable recovery guidance, achieving 65.37% diagnosis accuracy and 21.79% recovery rate on 257 cases.
-
CUDABeaver: Benchmarking LLM-Based Automated CUDA Debugging
CUDABEAVER benchmark and pass@k(M,C,A) metric show LLM CUDA debugging success drops by up to 40 percentage points under strict performance requirements.
-
SOCIA-EVO: Automated Simulator Construction via Dual-Anchored Bi-Level Optimization
SOCIA-EVO generates statistically consistent simulators by separating structural refinement from parameter calibration via bi-level optimization and falsifying strategies through execution feedback in a Bayesian-weigh...
-
Beyond Fixed Tests: Repository-Level Issue Resolution as Coevolution of Code and Behavioral Constraints
Agent-CoEvo is a multi-agent LLM framework that coevolves code patches and test patches to resolve repository-level issues, outperforming fixed-test baselines on SWE-bench Lite and SWT-bench Lite.
-
Closed-Loop LLM Discovery of Non-Standard Channel Priors in Vision Models
Closed-loop LLM search with AST-generated examples discovers non-standard channel widths that improve vision model performance over initial architectures on CIFAR-100.
-
ReCode: Improving LLM-based Code Repair with Fine-Grained Retrieval-Augmented Generation
ReCode improves LLM-based code repair by combining algorithm-type prediction with dual-encoder retrieval of authentic bug-fix pairs, and introduces RACodeBench for evaluation.
-
LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code
LiveCodeBench collects 400 recent contest problems to create a contamination-free benchmark evaluating LLMs on code generation and related capabilities like self-repair and execution.
-
Large Language Models Cannot Self-Correct Reasoning Yet
LLMs cannot reliably self-correct reasoning mistakes using only their internal capabilities and often degrade in performance without external feedback.
-
HTMLCure: Turning Browser Experience into State Guided Repair for Interactive HTML
HTMLCure uses browser-executed interaction trajectories to diagnose and repair LLM HTML outputs, expanding 97K prompts into a 40K refined SFT set that lifts a 27B model to 50.6 on HTMLBench-400 and 81.2 on MiniAppBench.
-
Automated LTL Specification Generation from Industrial Aerospace Requirements
AeroReq2LTL automates LTL generation from industrial aerospace requirements via LLMs with a data dictionary and templates, achieving 85% precision and 88% recall on real data.
-
Bridging the Gap between User Intent and LLM: A Requirement Alignment Approach for Code Generation
REA-Coder improves LLM code generation by iteratively aligning requirements with model understanding and verifying outputs against the aligned spec.
-
Let's Revise Step-by-Step: A Unified Local Search Framework for Code Generation with LLMs
The abstract claims a new local search framework for code generation, but the manuscript body is a different mathematics paper.
-
How Many Tries Does It Take? Iterative Self-Repair in LLM Code Generation Across Model Scales and Benchmarks
Iterative self-repair improves LLM code pass rates by 4.9-17.1 pp on HumanEval and 16-30 pp on MBPP across seven models, with gains concentrated early and syntax errors easier to fix than logical ones.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.