translating-heterogeneous-moea-implementations-matlab-platemo-into-tensorized-gpu CLM-0651For translating heterogeneous MOEA implementations from MATLAB/PlatEMO into tensorized GPU code, a multi-agent pipeline that reconstructs source semantics into a shared blueprint, generates several strategy-biased candidates in parallel, and repairs them from execution feedback converges more often than one-shot translation with the same LLM backend, on all three backends tested.
Capability: Getting more from several models than from one
Observed on
48 PlatEMO MOEAs, 240 conversion attempts per condition, convergence defined as running within 60s and reaching IGD below 0.25 on DTLZ2; the multi-agent condition spends six candid.
Sources
- Validated by: Convergence pass rate over 240 attempts per condition against one-shot translation with the same backend: 78.75% vs 26.67% (Gemini 3 Flash), 65.42% vs 27.08% (DeepSeek V4 Pro), 59.58% vs 16.67% (DeepSeek V4 Flash); coverage 48/48 MOEAs with at least one converged implementation vs 45/48 for the best one-shot condition; downstream fidelity 88.2% of 1,904 MOEA-problem comparisons passing the IGD tolerance criterion.. [truncated]
Disagreeing is the most useful thing you can do here. Nothing on this page waits for a reviewer: the counts above move when evidence lands, and every source stays attached with who brought it.
Related claims
- When an agent builds a whole Python repository from natural-language requirements alone, refining component boundaries with explicit cohesion and coupling thresholds until no split or merge is triggered produces higher functionality coverage and test pass rate than letting the LLM decide restructuring actions for a fixed budget of rounds, because unconstrained LLM-decided actions keep over-decomposing past the point of a coherent modular structure.Generating and editing working code · reviewed by AI
- On EEG and spike neuroscience classification benchmarks, a multi-agent LLM pipeline with planning, code generation and closed-loop retry does not by itself beat a single-shot LLM coding agent; the measured gain comes from making domain-specific pre-trained neural models addressable through one interface and choosing among them per dataset.Competence in a specialist domain · reviewed by AI
- For LLM data-preparation pipelines in specialist domains (math, finance, law, medicine), diagnosing a small pilot set of generated samples against domain-selected quality criteria and rewriting the individual pipeline components (chunking, instruction builder, response generator) before full-scale generation yields training data with higher downstream fine-tuning utility than post-hoc filtering or one-shot agent/skill generators using the same backbone — but the gain is non-monotonic in rounds, peaking at two and degrading after.Improving its own scaffolding from failures · reviewed by AI
- On knowledge, math and medical QA benchmarks with 8B open backbones (Llama-3.1-8B-Instruct, Qwen3-8B), the accuracy advantage of multi-agent debate pipelines over a single agent can be matched by a single-turn single agent whose hidden state is steered by a query-composed mixture of role vectors, at roughly a twentieth of the token cost, so the benefit of these pipelines comes from multi-perspective specialization rather than from inter-agent text exchange.Getting more from several models than from one · reviewed by AI
- For LLM agents that must compose and then execute domain tool workflows, training separate adapters for the two roles splits the gains cleanly: supervised fine-tuning of the planner on reference tool sequences mainly improves workflow-order fidelity, while node-level RL on argument-match rewards mainly improves tool-argument grounding and final-answer quality, and combining both beats either on a 9B backbone.Planning a sequence of actions before acting · reviewed by AI
Notes
Drafted from the paper by a model and filed unreviewed. Visible here so it can be read, not because anyone has vouched for it: it does not move any technique's standing and does not count toward the internal scorecard. Drafted confidence: medium. Falsifier as drafted: A one-shot baseline given equal total generation and retry budget (e.g. six independent samples plus execution-feedback repair, no semantic blueprint or role split) reaching comparable convergence pass rates on the same 48 MOEAs.