Pith. sign in

REVIEW 31 cited by

Magicoder: Empowering Code Generation with OSS-Instruct

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

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

arxiv 2312.02120 v2 pith:3J4VEV6A submitted 2023-12-04 cs.CL cs.AIcs.SE

Yuxiang Wei , Zhe Wang , Jiawei Liu , Yifeng Ding , Lingming Zhang This is my paper · ORCID
classification cs.CLcs.AIcs.SE
keywords codedatamagicodermodelsopen-sourceoss-instructinstructionllms
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

We introduce Magicoder, a series of fully open-source (code, weights, and data) Large Language Models (LLMs) for code that significantly closes the gap with top code models while having no more than 7B parameters. Magicoder models are trained on 75K synthetic instruction data using OSS-Instruct, a novel approach to enlightening LLMs with open-source code snippets to generate diverse instruction data for code. Our main motivation is to mitigate the inherent bias of the synthetic data generated by LLMs through the wealth of open-source references for the production of more realistic and controllable data. The orthogonality of OSS-Instruct and other data generation methods like Evol-Instruct further enables us to build an enhanced MagicoderS. Both Magicoder and MagicoderS substantially outperform state-of-the-art code models with similar or even larger sizes on a wide range of coding benchmarks. Notably, MagicoderS-CL-7B based on CodeLlama even surpasses the prominent ChatGPT on HumanEval+ (66.5 vs. 65.9 in pass@1 ). Overall, OSS-Instruct opens a new direction for crafting diverse synthetic instruction data for code using abundant open-source references.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 31 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. rStar2-Agent: Agentic Reasoning Technical Report

    cs.CL 2025-08 conditional novelty 7.0 of 10

    A 14B model trained with agentic RL and a resample-on-correct rollout strategy scores 80.6% on AIME24 and 69.8% on AIME25, nearly matching DeepSeek-R1 (671B) in one week on 64 GPUs.

  2. SWE-Flow: Synthesizing Software Engineering Data in a Test-Driven Manner

    cs.CL 2025-06 conditional novelty 7.0 of 10

    SWE-Flow synthesizes incremental, test-driven development tasks from real GitHub projects and shows that fine-tuning Qwen2.5-Coder-32B-Instruct on them improves performance on the resulting SWE-Flow-Bench benchmark.

  3. COFFE: A Code Efficiency Benchmark for Code Generation

    cs.SE 2025-02 conditional novelty 7.0 of 10

    A new benchmark, COFFE, uses stressful test cases and CPU instruction counts to show LLM-generated code is often correct but time-inefficient.

  4. LEAP: Lean Environment-Feedback via Adaptive Pruning for Code RL in GPU Kernel Generation

    cs.LG 2026-08 conditional novelty 6.0 of 10

    LEAP combines difficulty-conditioned pruning with pairwise rank-based rewards to improve multi-turn reinforcement learning for CUDA kernel generation.

  5. MoE$^2$-LoRA: When MoE Models Meet MoE-style Low-Rank Adaptation

    cs.CL 2026-07 conditional novelty 6.0 of 10

    Routing LoRA adapters with the frozen base router's logits plus a shared cross-layer adapter pool gives the best PEFT accuracy and retention on three MoE backbones.

  6. SciCodePile: A 128GB Corpus and Executable Benchmark for Challenging Scientific Code Generation

    cs.SE 2026-07 conditional novelty 6.0 of 10

    SciCodePile contributes a 128GB scientific code corpus and a 200-task executable benchmark showing that current LLMs still fall far short of reliable scientific code generation.

  7. Function-Aware Fill-in-the-Middle as Mid-Training for Coding Agent Foundation Models

    cs.AI 2026-07 unverdicted novelty 6.0 of 10

    Mid-training on function-aware fill-in-the-middle data improves SWE-Bench by 2.8-5.4 points and reduces post-training capability loss.

  8. Are we Merging the Right Models? Impact of Expert Training Duration on Model Merging for LLMs

    cs.LG 2026-07 unverdicted novelty 6.0 of 10

    The optimal training duration for expert models before merging depends on the merging method; sparsification-based merging benefits from overtrained experts.

  9. LP-SFT: Local-Preserving Supervised Fine-Tuning via Multimodal Entropy Structure

    cs.CL 2026-07 conditional novelty 6.0 of 10

    Preserving adaptive non-label local structure from the base model during SFT improves the pass@1 vs pass@k trade-off and reduces catastrophic forgetting versus vanilla cross-entropy and recent SFT variants.

  10. GLAN-QnA-KR: A Seedless Taxonomy-Driven Korean Instruction Corpus

    cs.CL 2026-05 conditional novelty 6.0 of 10

    A 303,581-row Korean instruction corpus generated seedlessly from a 1,084-discipline taxonomy, with near-zero duplicates and low measured overlap with KMMLU, KoBEST, and HAE-RAE-Bench.

  11. Modeling Code: Is Text All You Need?

    cs.AI 2025-07 conditional novelty 6.0 of 10

    A GNN-encoded LLVM IR graph, prepended as soft prompts to a frozen code LLM, improves accuracy on device mapping, algorithm classification, vulnerability detection, and code translation tasks.

  12. Turning the Tide: Repository-based Code Reflection

    cs.SE 2025-07 conditional novelty 6.0 of 10

    A new 1,888-case, six-language benchmark and a trained model measure and improve how code LLMs edit and repair code inside multi-file repositories.

  13. BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity

    cs.LG 2025-07 conditional novelty 6.0 of 10

    A ReLU-routed MoE with chunk-level sparsity training objectives and custom kernels combining activation sparsity with speculative decoding achieves over 70% 8-token chunk sparsity and up to 3.67x end-side speedup.

  14. Multilingual Multimodal Software Developer for Code Generation

    cs.CL 2025-07 conditional novelty 6.0 of 10

    A 7B vision-language model trained on synthetic diagram-to-code data outperforms several larger open-weight models on a new 10-language UML/flowchart code-generation benchmark.

  15. Does Learning Mathematical Problem-Solving Generalize to Broader Reasoning?

    cs.CL 2025-07 conditional novelty 6.0 of 10

    Long chain-of-thought and RL training on math problems improves general reasoning benchmarks, while short chain-of-thought math fine-tuning often degrades performance.

  16. A Hierarchical and Evolvable Benchmark for Fine-Grained Code Instruction Following with Multi-Turn Feedback

    cs.SE 2025-07 conditional novelty 6.0 of 10

    MultiCodeIF introduces a 2,021-task, 14-language benchmark with 27 constraint types to evaluate code instruction following, finding that multi-level constraints sharply reduce model success and iterative feedback subs...

  17. ScaleRTL: Scaling LLMs with Reasoning Data and Test-Time Compute for Accurate RTL Code Generation

    cs.AR 2025-06 conditional novelty 6.0 of 10

    A reasoning LLM fine-tuned on 3.5B tokens of RTL chain-of-thought data, with iterative test-time self-correction, sets new state-of-the-art results on VerilogEval and RTLLM.

  18. Seed-Coder: Let the Code Model Curate Data for Itself

    cs.CL 2025-06 conditional novelty 6.0 of 10

    Seed-Coder shows that an LLM-trained quality scorer can filter 6T tokens of code data and yield 8B models that outperform similar-size open code models.

  19. EXP-Bench: Can AI Conduct AI Research Experiments?

    cs.AI 2025-05 conditional novelty 6.0 of 10

    EXP-Bench is a new benchmark of 461 end-to-end AI research experiments, and leading AI agents complete fewer than 1 percent of them successfully.

  20. ExeSQL: Self-Taught Text-to-SQL Models with Execution-Driven Bootstrapping for SQL Dialects

    cs.CL 2025-05 conditional novelty 6.0 of 10

    Execution-driven bootstrapping, where a model generates SQL, executes it, and keeps only queries that run, lets a 7B model outperform GPT-4o on PostgreSQL, MySQL, and Oracle text-to-SQL benchmarks.

  21. CSR-Bench: Benchmarking LLM Agents in Deployment of Computer Science Research Repositories

    cs.SE 2025-02 conditional novelty 6.0 of 10

    CSR-Bench and CSR-Agents show that LLM agents can complete under half of setup and data-download steps, and between 15 and 29 percent of training, inference, and evaluation steps, on 100 research repositories.

  22. Route-Align-Verify for Functional Correctness in Code Generation

    cs.SE 2026-08 reject novelty 5.0 of 10

    RAV combines task-aware prompt routing, aligned LoRA fine-tuning, and public-test-based candidate selection to improve MBPP pass@1, but its verification step leaks evaluation tests.

  23. Quantize with Confidence? An Empirical Study of Quantization for Code Generation

    cs.SE 2026-07 conditional novelty 5.0 of 10

    4-bit quantization of code LLMs mostly preserves pass@1, but AQLM matches or beats full precision while QuIP# degrades most on complex prompts; sensitivity is model-dependent.

  24. A Survey on Evaluating Quality and Trustworthiness in LLM-Generated Data

    cs.AI 2026-01 conditional novelty 5.0 of 10

    A metric-oriented survey that classifies intrinsic quality and trustworthiness metrics for LLM-generated data across six modalities and documents systematic evaluation gaps in the current literature.

  25. SelfAug: Mitigating Catastrophic Forgetting in Retrieval-Augmented Generation via Distribution Self-Alignment

    cs.CL 2025-09 conditional novelty 5.0 of 10

    Adding a KL penalty between fine-tuned and original model logits on input tokens during RAG fine-tuning reduces catastrophic forgetting while preserving downstream performance.

  26. Grove MoE: Towards Efficient and Superior MoE LLMs with Adjugate Experts

    cs.CL 2025-08 unverdicted novelty 5.0 of 10

    Grove MoE uses unequal-size adjugate experts with complexity-based activation to run 33B-parameter models at roughly 3.1 to 3.3B active parameters while matching larger open models in benchmarks.

  27. ALPS: Attention Localization and Pruning Strategy for Efficient Alignment of Large Language Models

    cs.CL 2025-05 conditional novelty 5.0 of 10

    ALPS selects task-sensitive attention heads by measuring Wasserstein distance between base and task-tuned weights, and freezing other heads during fine-tuning improves performance and efficiency.

  28. Multi-Agent Collaboration for Multilingual Code Instruction Tuning

    cs.CL 2025-02 reject novelty 5.0 of 10

    A multi-agent collaboration framework for generating multilingual code instruction data is introduced, but the reported evaluations show the fine-tuned model underperforming its own base instruct model.

  29. Training-Free versus Training-Based Intent Classification in LLMs: Accuracy, Robustness, and Failure Modes

    cs.CL 2026-08 conditional novelty 4.0 of 10

    Statistical classifiers built on LLM activation norms and coordinates match or beat trained MLP heads on coarse intent routing and resist camouflage better, while MLPs win on fine-grained subfield distinctions.

  30. Position: Intelligent Coding Systems Should Write Programs with Justifications

    cs.SE 2025-08 conditional novelty 4.0 of 10

    A position paper advocating that intelligent coding systems should accompany code with justified explanations that are cognitively aligned and semantically faithful.

  31. Infinite-Instruct: Synthesizing Scaling Code instruction Data with Bidirectional Synthesis and Static Verification

    cs.CL 2025-05 reject novelty 4.0 of 10

    A synthetic data pipeline for code instruction tuning reports large benchmark gains, but the headline improvements are internally inconsistent.

Pith tools