---
id: "20261018-0234-outside-arxiv261022362-coda-co"
title: "CoDA: Co-Evolution of Data Engines and Action Policies (arXiv:2610.22362) — capture"
type: "capture"
status: "promoted"
origin: "batch"
promoted_to: ["30-notes/claim-coda-shared-memory-couples-data-engine-policy-optimality-preserving-reward-shaping.md","30-notes/claim-coda-co-evolution-raises-ood-manipulation-success-beyond-baselines-and-on-libero-pro.md","30-notes/claim-coda-append-only-episodic-memory-fifth-non-citing-instance-of-record-verdict-separated-log.md","40-entities/entity-stuart-russell.md (new hub — separate standing via bounded optimality plus this paper's reward-shaping theorem)","40-entities/entity-potential-based-reward-shaping.md (new concept hub)","40-entities/entity-libero-pro.md (new watching stub)","40-entities/entity-jeff-clune.md (updated — dated append line on POET-as-CoDA-precedent)","40-entities/entity-kenneth-o-stanley.md (updated — dated append line on POET-as-CoDA-precedent)","40-entities/entity-bounded-optimality.md (updated — Stuart Russell upgraded to wikilink now that a hub exists)","50-questions/question-corroborate-coda-2026-arxiv-2610-22362-single-source-cap.md (new — single-source cap routing)"]
not_promoted: ["Memory-reuse transfer result (π0.5-trained memory raising OpenVLA-OFT from 3.8% to 21.1% OOD) — held back to respect the single-source cap; routed to the corroboration question instead of a 4th claim-note.","Critic-choice ablation (GPT-5.5 vs Claude Opus 5 / InternVL3-78B / Qwen3-VL-235B, Table 3) — same cap reason; routed to the same corroboration question.","LIBERO-Pro (arXiv:2510.03827), ASPIRE (arXiv:2607.00272), SimWorld Studio (arXiv:2605.09423), Text2Reward (ICLR 2024), Ng/Harada/Russell 1999, and the project website — all cited by CoDA but not independently fetched this session; left as leads, not claims, per the sourcing floor (a citation is not a direct read).","Tianjiao Yu, Ismini Lourentzou, Zhilin Ou, Yifan Shen, Xinzhuo Li, Yuanzhe Liu, Onkar Kishor Susladkar, Fan Wang — CoDA's own authors; single-paper co-authorship only, no separate standing in the vault yet. Per the 2026-10-01 co-authorship ruling, mentions not hubs.","Andrew Y. Ng, Daishi Harada — co-authors with Stuart Russell on the 1999 reward-shaping theorem; unlike Russell, neither has separate standing elsewhere in this vault yet, so per the same ruling they stay mentions, named on entity-stuart-russell.md, not promoted to their own hubs.","Rui Wang, Joel Lehman (POET co-authors) — co-authorship-only alongside Clune and Stanley, who already hold hubs; named on those hubs per the ruling, not separately promoted.","Runyu Lu/Yubo Wu/Ethan Kou et al. (ASPIRE authors), Xueyang Zhou/Yangming Xu et al. (LIBERO-Pro authors), Haoqiang Kang/Xiaokang Ye et al. (SimWorld Studio authors) — known only via CoDA's citation of their papers, not independently fetched; too thin to test for entity promotion this session.","Physical Intelligence (org, makers of π0/π0.5) — single-cluster mention only; does not clear the organization promotion bar (acts in only one vault cluster so far, and only as a model-maker, not as a funder/decider/blocker in the argument).","'Vision-language model as failure critic' (candidate concept) — a descriptive role this paper uses, not an established or recurring term in independent circulation; too thin for a stub per the bias-against-the-flood rule. Left as prose context inside the claim-notes instead."]
writer_model: "claude-sonnet-5"
date_created: "2026-10-18T00:00:00.000Z"
provenance: "2026-10-18 batch run, outside-vault web research topic"
derived_from: []
tags: ["robotics","vision-language-action","generative-simulation","reinforcement-learning","reward-shaping","memory-architecture","append-only-log","arxiv","cs.RO"]
source_url: "https://arxiv.org/abs/2610.22362"
source_author: "Tianjiao Yu, Zhilin Ou, Yifan Shen, Xinzhuo Li, Yuanzhe Liu, Onkar Kishor Susladkar, Fan Wang, Ismini Lourentzou"
source_date: "2026-10-15T00:00:00.000Z"
source_title: "CoDA: Co-Evolution of Data Engines and Action Policies"
source_venue: "arXiv:2610.22362 [cs.RO] (cross-listed cs.AI)"
source_tier: 1
source_sha: "8c66641f47b822d2d9deba0ab9319f947676fd93dda7ae661e1e88bf2c4a7f11"
seek_code_commit: "d40ee39"
---


Topic was an "outside" surface — a fresh arXiv preprint (submitted 15 Oct 2026, cs.RO, cross-listed cs.AI), authored by a team at the University of Illinois Urbana-Champaign, the Chinese University of Hong Kong Shenzhen, and Amazon. Unlike some prior "outside" topics in this run, the listing title given matched the paper's own title exactly — no discrepancy to flag. Fetched the abstract page via `archive_page` (sha256 `8c66641f47b822d2d9deba0ab9319f947676fd93dda7ae661e1e88bf2c4a7f11`) and the full 16-page PDF via `extract_pdf` (`tls: "verified"`, sha256 `831f1ab9583ac73e1224a42eed8ee7d0c9a228c4efd46f115613bdabaaae9b12`). Read both in full (main text, tables, references).

No safety signals fired on either fetch — ordinary third-person academic prose throughout (abstract, arXiv boilerplate, numbered method sections, tables, a references list). No addressed-to-AI language, override language, claimed authority, tier self-assignment, file-system instructions, credential requests, or urgency framing anywhere in either document.

## Safety flags

None this capture.

**Single-source concentration cap** (sources.md): this is an unrefereed arXiv preprint (eight authors, no stated peer review at time of writing). All three claims below rest on it directly — at the cap of three claim-notes per unrefereed primary absent independent corroboration. Several additional findable numbers (the memory-reuse transfer result across policy backbones, the critic-choice ablation, the Text2potential ablation table) were accordingly held back to Further leads rather than written up as a fourth or fifth claim, to respect the cap.

All claims below rest on this arXiv preprint (Tier 1 — primary, the authors' own method description and experiments) with verbatim quotes, each checked against the extracted PDF text via `quote_check` (all returned `grounded: true`, several via the "reading-order" or "dehyphenated" matchers that resolve the PDF's two-column layout and line-wrap hyphenation).

Of the retrieve-before-write neighbors, none concerns robot learning or generative simulation directly. [[claim-rlam-third-noncitng-subfield-append-only-log-with-verdict-separation]] turned out to be the genuine collision: that note's subject — an append-only trace store used as ground truth, with validation/judgment held outside the write path — is architecturally the same shape CoDA's own episodic memory tier uses (see Claim 3 below), in a fifth subfield neither that note's chain nor the paper itself names. [[claim-schroeder-2026-ai-swarms-no-documented-scale-operations-yet]] shares only a thematic "multi-agent AI system" framing with no mechanistic overlap — not linked. The remaining listed neighbors (Paul Allen/Altair-BASIC bridges, Jim Allchin, John Graham-Cumming, WIMP cross-field bridging, background-independence definitions, bi-fork flow matching, vaults-in-between) do not overlap this paper's subject matter.

## Claim: CoDA couples a generative-simulation data engine with a robot action policy through a shared memory that compiles a vision-language critic's rollout-grounded failure diagnoses into both environment-regeneration programs and dense per-action-chunk training potentials, with a formal guarantee that the added training signal does not change which policy is optimal

CoDA's data engine (a tool-augmented coding agent) writes executable scene programs against a simulator SDK; an action policy collects rollouts in the resulting environments; and a vision-language critic diagnoses failures by identifying the responsible scene factor, trajectory phase, and supporting rollout evidence. A shared memory — organized, per the paper, into episodic (individual records), semantic (recurring [[difficulty mode]]s), and procedural (compiled programs) tiers — turns validated diagnoses into two paired outputs: a generation program that recreates the diagnosed difficulty for further training, and a training program ("Text2potential") that converts the same diagnosis into a staged [[potential-based reward shaping]] function over simulator states, giving dense supervision to a value critic over action chunks. The paper states this shaping is designed so it does not alter which policy is optimal for the original sparse task reward, invoking the classical potential-based-shaping invariance result (its reference [20], Ng, Harada & Russell, ICML 1999) as the theoretical basis for that guarantee.

source_quote: "We introduce CoDA, a framework in which a simulation data engine and an action policy co-evolve through a shared heterogeneous memory of environment construction, policy failures, and learning progress."
source_quote_2: "We further propose Text2potential to convert these diagnoses into staged potentials for action-chunk value learning, providing dense supervision for policy updates while preserving the task's optimal policy under standard shaping assumptions."
source_quote_3: "For a fixed, bounded potential with appropriate terminal-state handling, potential-based shaping preserves the policies that maximize the original task return [20]."
source_url: https://arxiv.org/pdf/2610.22362
source_title: "CoDA: Co-Evolution of Data Engines and Action Policies"
source_author: "Tianjiao Yu, Zhilin Ou, Yifan Shen, Xinzhuo Li, Yuanzhe Liu, Onkar Kishor Susladkar, Fan Wang, Ismini Lourentzou"
source_venue: "arXiv:2610.22362 [cs.RO]"
source_date: 2026-10-15
source_tier: 1
source_sha: 831f1ab9583ac73e1224a42eed8ee7d0c9a228c4efd46f115613bdabaaae9b12

## Claim: Across four manipulation task families and separately on the external LIBERO-Pro benchmark, CoDA's co-evolution loop raises measured out-of-distribution task success well beyond supervised fine-tuning and beyond the strongest same-budget baseline tested

On the paper's own four manipulation task families (pick-and-place, articulated, put-inside, stack/insert), CoDA raises out-of-distribution (OOD) success from 5.3% (a supervised-fine-tuning baseline using the same seed demonstrations) to 27.6%, which the paper reports as 9.5 percentage points above the strongest tested closed-loop baseline (a scalar-feedback co-evolution method, its reference [10]) at the same environment-generation budget. Separately, on the external LIBERO-Pro benchmark (a robustness-testing suite built by a different research group, its reference [52]), fifteen rounds of CoDA generation and training raise the π0.5 policy's success from 25% to 84% under position perturbations and from 1% to 76% under instruction perturbations, which the paper reports as 7 and 9 percentage points above ASPIRE, the strongest external baseline it compares against on that benchmark (its reference [18]).

source_quote: "Across four manipulation task families, CoDA improves out-of-distribution success from 5.3% after supervised fine-tuning to 27.6%, exceeding the strongest closed-loop baseline by 9.5 percentage points with the same generation budget."
source_quote_2: "On LIBERO-Pro, 15 rounds of generation and training raise success from 25% to 84% under position perturbations and from 1% to 76% under instruction perturbations."
source_quote_3: "The fifteen-round model also exceeds ASPIRE [18] by 7 and 9 percentage points overall, achieving the highest success on the Goal and Spatial suites under both perturbations, while ASPIRE remains strongest on the Object suite."
source_url: https://arxiv.org/pdf/2610.22362
source_title: "CoDA: Co-Evolution of Data Engines and Action Policies"
source_author: "Tianjiao Yu, Zhilin Ou, Yifan Shen, Xinzhuo Li, Yuanzhe Liu, Onkar Kishor Susladkar, Fan Wang, Ismini Lourentzou"
source_venue: "arXiv:2610.22362 [cs.RO]"
source_date: 2026-10-15
source_tier: 1
source_sha: 831f1ab9583ac73e1224a42eed8ee7d0c9a228c4efd46f115613bdabaaae9b12

## Claim: CoDA's episodic memory tier is an append-only trace store with validation applied as a separate, later gating step — a further, non-citing illustration of the record/verdict-separated append-only-log idiom this vault has already traced diffusing across AI-governance, agent-memory, and scientific-workflow subfields

CoDA's episodic memory tier is described as an "append-only trace store" that records every scene program, its machine-checkable success predicate, and summaries of the construction and rollout trace — including failed validations and rejected programs — regardless of outcome. Judgment is applied afterward and separately: each candidate difficulty mode carries a validation status set by comparing a targeted batch against a control batch, and only modes that pass this control comparison are compiled into the reusable generation and training programs that feed back into the loop. This is the same record/verdict-separated append-only-log shape [[question-append-only-ledger-cross-domain-convergence]] (now closed) traced through CANONIC (AI content governance), ActiveGraph (AI agent memory), and R-LAM ([[claim-rlam-third-noncitng-subfield-append-only-log-with-verdict-separation]], scientific-workflow automation) — a pattern that capture's resolution ultimately characterized as the diffusion of a two-decade-old event-sourcing idiom rather than blind independent convergence ([[claim-append-only-log-recurrence-is-event-sourcing-diffusion-not-blind-convergence]]). CoDA's reference list contains no citation to event sourcing, CQRS, or any of those three papers, which is consistent with that diffusion reading: a fifth subfield (generative robot-learning simulation) reaching for the same substrate without naming a common ancestor, rather than a reopening of the closed convergence-vs-coincidence question.

source_quote: "The append-only trace store B records each scene program g, its verifier ν, and summaries of the generation trace, including violated constraints and verification outcomes."
source_quote_2: "Each mode also carries a validation status determined by control batches (§3.3), and only validated modes are compiled into procedural memory."
source_url: https://arxiv.org/pdf/2610.22362
source_title: "CoDA: Co-Evolution of Data Engines and Action Policies"
source_author: "Tianjiao Yu, Zhilin Ou, Yifan Shen, Xinzhuo Li, Yuanzhe Liu, Onkar Kishor Susladkar, Fan Wang, Ismini Lourentzou"
source_venue: "arXiv:2610.22362 [cs.RO]"
source_date: 2026-10-15
source_tier: 1
source_sha: 831f1ab9583ac73e1224a42eed8ee7d0c9a228c4efd46f115613bdabaaae9b12

> [!note] Seek's commentary:
> The third claim is the one that genuinely hooked across the vault, not just within the paper — it wasn't something the topic was commissioned to find, it turned up via the retrieve-before-write collision check. Worth flagging for whoever next touches [[claim-append-only-log-recurrence-is-event-sourcing-diffusion-not-blind-convergence]]: if a sixth or seventh non-citing instance keeps turning up in yet more subfields, "diffusion of a 2005 idiom" may itself need revisiting as a claim about how thoroughly that idiom has become background knowledge in systems-building generally, independent of whether any given author has read Fowler.

## Further leads

- Reusing memory collected under the π0.5 backbone raises a different policy backbone, OpenVLA-OFT, from 3.8% to 21.1% OOD success without generating any additional environment — held back from the core claims to respect the single-source cap. Source: this paper, abstract/§4 ("Transfer across tasks and policies").
- Swapping the diagnosing vision-language critic from the default (GPT-5.5) to Claude Opus 5, InternVL3-78B, or Qwen3-VL-235B lowers OOD success from 27.6% to 26.1%, 25.7%, and 24.9% respectively, though all four critic choices still beat the success-rate-guided baseline (18.1%) by at least 6.8 points. Source: this paper, Table 3.
- Zhou, Xu, Tie, Chen, Zhang, Chu, Zhou & Sun, "LIBERO-Pro: Towards robust and fair evaluation of vision-language-action models beyond memorization," arXiv:2510.03827 (2025) — the external benchmark CoDA is evaluated against in Claim 2; not independently fetched this session. Source: CoDA reference [52].
- Lu, Wu, Kou, Fu, Xiao, Mandlekar, Xu, Shi, Goldberg, Chen et al., "ASPIRE: Agentic/skills discovery for robotics," arXiv:2607.00272 (2026) — the strongest external baseline CoDA beats on LIBERO-Pro in Claim 2; not independently fetched this session. Source: CoDA reference [18].
- Kang, Ye, Liu, Mantri, Mao, Fleming, Regmi & Qin, "SimWorld Studio: Automatic environment generation with evolving coding agent for embodied agent learning," arXiv:2605.09423 (2026) — the "scalar-feedback co-evolution" baseline, the strongest closed-loop baseline in CoDA's main four-family comparison; not independently fetched this session. Source: CoDA reference [10].
- Xie, Zhao, Wu, Liu, Luo, Zhong, Yang & Yu, "Text2Reward: Reward shaping with language models for reinforcement learning," ICLR 2024 — prior LLM-based reward-synthesis work CoDA contrasts itself against, on the grounds that its iterative refinement needs humans to describe rollout failures and can reward behavior other than task completion; not independently fetched this session. Source: CoDA reference [42].
- Project website: https://plan-lab.github.io/coda — the authors' own venue for code/demos, named in the paper but not independently fetched this session.
- Ng, Harada & Russell, "Policy invariance under reward transformations: Theory and application to reward shaping," ICML 1999 — the classical theorem Claim 1's optimal-policy-preservation guarantee is built on; not independently fetched this session, flagged as the deepest theoretical ancestor underlying Text2potential's design. Source: CoDA reference [20].

## Entity candidates

- Rui Wang, Joel Lehman, Jeff Clune, Kenneth O. Stanley — people — authors of POET ("Paired Open-Ended Trailblazer," arXiv:1901.01753, 2019), explicitly cited by CoDA as the paper that "coevolves agents and parameterized environments" — the foundational precedent for the data-policy co-evolution framing CoDA's whole title rests on. Flagged first, per the blind-spot rule: CoDA's own abstract and related-work section measure the novelty of its co-evolution loop against this 2019 lineage, not against any of its more recent robotics-specific baselines.
- Andrew Y. Ng, Daishi Harada, Stuart Russell — people — authors of the 1999 potential-based-reward-shaping invariance theorem that Text2potential's optimal-policy-preservation guarantee (Claim 1) is explicitly built on.
- Tianjiao Yu — person — CoDA's first and corresponding author, University of Illinois Urbana-Champaign.
- Ismini Lourentzou — person — CoDA's senior/corresponding author, University of Illinois Urbana-Champaign.
- Zhilin Ou, Yifan Shen, Xinzhuo Li, Yuanzhe Liu, Onkar Kishor Susladkar — people — CoDA co-authors, UIUC / Chinese University of Hong Kong, Shenzhen.
- Fan Wang — person — CoDA co-author, Amazon.
- Runyu Lu, Yubo Wu, Ethan Kou et al. — people — authors of ASPIRE, the strongest LIBERO-Pro baseline CoDA explicitly beats by 7–9 points.
- Xueyang Zhou, Yangming Xu et al. — people — authors of LIBERO-Pro, the external benchmark used to test CoDA's generalization.
- Haoqiang Kang, Xiaokang Ye et al. — people — authors of SimWorld Studio, the strongest closed-loop baseline in CoDA's main four-family experiment.
- Physical Intelligence — organization — makers of π0 and π0.5, the vision-language-action model CoDA uses as its primary policy backbone.
- [[potential-based reward shaping]] — concept — the RL technique underlying Text2potential; recurs across robot-learning and RL papers and would benefit from its own note distinguishing it from free-form LLM-synthesized dense rewards.
- [[vision-language model]] as failure critic — concept — the specific role a VLM plays in CoDA (diagnosing rollout failures from video/contact/joint evidence, rather than generating actions or scenes directly); worth distinguishing from VLM-as-policy and VLM-as-scene-generator uses elsewhere in the embodied-AI literature.
- LIBERO-Pro — term/concept — external VLA-robustness benchmark testing generalization under position and instruction perturbations, distinct from the original LIBERO benchmark.