{"id":"dc5358bc-f28c-4dd2-9ab3-7ac703bd5fe5","arxiv_id":"2609.36630","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"A survey defining Agent Distillation and organizing agent-to-agent knowledge transfer into model, artifact, and harness substrates with a causal evaluation framework.","lead":"This survey defines Agent Distillation as the persistent transfer of task-solving knowledge from a teacher agent to a student agent, and organizes the literature by where that knowledge is kept: in the model parameters, in reusable artifacts, or in the execution harness. Read it for a common vocabulary and an evaluation framework that could make the many scattered agent-distillation papers comparable.","discovery_kind":"review","skeptic_critique":{"model":"deepseek/deepseek-v4-flash","headline":"The taxonomy's classification rule is non-operational for learned-control components: §3's functional criterion collides with §6.2/§5.4 assigning learned routers and retrievers to M_S, so single-substrate cell assignments may be irreproducible; an inter-coder reliability check would settle it.","rationale":"The reader's weakest_assumption flagged that methods may resist single-primary-substrate assignment, weakening the taxonomy's organizing power. My concern is related but more specific: the failure is not only empirical overlap but an internal inconsistency in the classification rule itself. §3 declares substrate determined by how knowledge participates in future task solving, not by physical format, yet §6.2/§5.4/§7.2 assign learned control components (routers, retrievers, workflow generators) to M_S by virtue of living in neural weights, reserving H_S for surrounding runtime. A state-conditioned selection policy is harness knowledge by §3's own definition, so the same transferred knowledge receives M or H depending on the implementation the original paper chose — a physical-format judgment the survey disclaims. This is an internal-consistency concern (rule contradicts its own application), not merely disagreement with a broader consensus. I kept the verdict UNCHANGED for three reasons. First, the survey is heavily hedged: it explicitly stapled 'for organizational purposes' to the classification, admits in §9.1, §7.1, and §6.4 that substrate attribution is an identification problem, and demands source-agnostic labels when access is unavailable — the authors are not overclaiming. Second, the paper's most durable contributions, the scope definition (persistent teacher-derived knowledge incorporation) and the §8 evaluation framework (claim tuples, ablation, intervention, Eq 26–28), survive even if the single-substrate cells are heuristic; the evaluation section is explicitly substrate-agnostic in its caution. Third, the concern is testable and, if confirmed, would weaken the novelty of the substrate organiization but would not overturn acceptance of a careful survey that already flags the limits. My check is designed to measure the actual reproducibility cost: an inter-coder study applying the §3 rule blind, plus a direct consistency probe on a learned router. If the check showed kappa below roughly 0.5 with disagreements concentrated on learned-control methods, I would move to CONDITIONAL and require the authors to mark single-substrate assignments as interpretive or provide the coding protocol, since the taxonomy's organizing contribution — the claimed first unified definition plus substrate classification — would then rest on the authors' privileged reading. Short of that evidence, the current ACCEPT at MODERATE confidence is the right verdict, and my concern does not change it.","tokens_in":42208,"tokens_out":7256,"duration_ms":56311,"concrete_test":"Take roughly 30 named core methods spanning §4–§6 (e.g., FireAct, AgentBank, Agent Memory Distillation, SKILL-KD, FlowScout, RouteLLM, AFlow, GPTSwarm, Agent Harness Distillation, Kang et al. tool-using models). Give two independent annotators only the original cited papers plus the §3 rule and Eq 21, and ask each to assign a primary substrate (M/R/H) and hybrid status. Compute Cohen's kappa on primary substrate, then inspect whether disagreement concentrates on learned-router, learned-retriever, or learned-workflow methods. Independently, apply the §3 rule and §6.2's treatment to one learned router and show the two readings yield M versus H.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing engine of the survey is reproducibility of classifier assignment: the claim (contribution 2, §3, 3×3 matrix of §7.1, hybrid criterion of Eq 21) that every core method admits a primary substrate (M/R/H) under the §3 rule, 'substrate membership is determined by how the retained knowledge participates in future task solving, rather than by its physical format or the mechanism used to obtain it.' That rule is non-operational for a class central to the agentic literature: learned control components. §6.2 assigns a learned router (RouteLLM) to M_S ('if the transferable knowledge is retained in the parameters of a learned router, that router is a parametric decision model and the primary update belongs to M_S'), reserving H_S for 'the surrounding runtime that queries the router.' §5.4 does the same for learned retrievers; §7.2 for learned workflow generation. Yet under §3's own definition ('knowledge that governs the sequencing and control of execution belongs to the harness'), a router that selects models or artifacts is harness knowledge. So the same decision function is classified parametric when implemented as learned weights and harness when written as explicit rules or a lookup table — precisely the physical-format criterion §3 disavows. The paper acknowledges the underlying problem (§9.1: 'Different combinations of M, R, and H can produce the same final behavior'; §7.1 demands source-agnostic labels; §6.4 refuses a single label for joint changes), yet Sections 4–6 and Table 1 still issue single-substrate, single-cell assignments for the surveyed corpus (b254, b379, and the RouteLLM and Agent Memory Distillation placements). Those assignments are made by reading published method descriptions, which rarely document the functional-participation detail the rule requires. The taxonomy's organizing power therefore rests on the authors' privileged reading rather than on a criterion an independent coder can apply consistently.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek/deepseek-v4-flash","summary":"The paper defines Agent Distillation as the persistent transfer of task-solving knowledge from a teacher agent to a student agent, and organizes the field by knowledge substrate: parametric model M, reusable artifacts R, and execution harness H. It provides a definitional formalization (Eqs 1-9 for agent realization, competence, and the distillation objective; Eqs 16-21 for harness distillation, cross-substrate translation, and the hybrid criterion), a 3x3 source-destination matrix (Eq 18), an evaluation framework that separates retention from causal contribution and deployed utility (Eqs 22-28), a benchmark landscape (Table 3), and open challenges. Scope boundaries are drawn carefully and hedged: in-context learning, inference-time consultation, and transient feedback are excluded when their effects end with the episode, while self-evolution is included when prior experience supplies teacher-side evidence. The survey is heavily hedged throughout, with explicit acknowledgment in 7.1 that the matrix is a conceptual accounting device, in 6.4 that joint changes should not be forced into a single label, and in 9.1 that different (M,R,H) combinations can produce the same behavior.","tokens_in":42597,"tokens_out":7900,"duration_ms":57473,"significance":"If the framework holds, the paper fills a genuine gap: distillation surveys are model-centered while agent surveys do not organize around persistent teacher-student transfer, and the substrate taxonomy gives the field a shared vocabulary and an evaluation protocol (claim tuple, substrate ablations, synergy measure) that is more rigorous than typical survey evaluation discussion. The formalization is definitional rather than empirical, but it is internally consistent, and the scope boundaries (2.3) are explicit enough to be falsifiable as classification rules. The evaluation framework in 8.1-8.2, particularly the claim tuple in Eq 22, the marginal-contribution and synergy measures in Eqs 27-28, and the benchmark caveats in Table 3, is a substantive contribution that goes beyond what most roadmaps offer. The main risk is the operationality of the substrate classification rule for learned control components, where the paper's own sections apply the rule inconsistently; this bears directly on contributions (2) and (3), so it needs to be resolved before the taxonomy can serve as the field's organizing dimension.","major_comments":[{"comment":"The substrate classification rule is applied inconsistently to learned control components, which is load-bearing for contribution (2). Section 3 states that 'knowledge that governs the sequencing and control of execution belongs to the harness' and that membership is determined by how knowledge participates in future task solving 'rather than by its physical format.' Yet 6.2 assigns a learned router (RouteLLM) to M_S because the knowledge is 'retained in the parameters,' reserving H_S for the runtime that queries it, and 5.4 assigns a learned retriever to M_S while placing the 'runtime policy that decides when to search' in H_S. Section 6.2 then states that for tool and artifact routing, 'the selection and invocation policy resides in H_S.' Under the 3 functional rule, a learned router or learned artifact-retrieval policy governs which component executes and therefore belongs to the harness; the 5.4/6.2 assignments instead privilege the retention locus (parameters), which is precisely the physical-format criterion 3 disavows. Moreover, the two sections disagree about where learned artifact-selection knowledge lives: 5.4 puts the learned retriever in M_S while 6.2 puts the artifact-selection policy in H_S. Since the 3x3 matrix (Eq 18) and the hybrid criterion (Eq 21) presuppose a reproducible primary-substrate assignment, the paper should state an explicit reconciliation rule (e.g., learned decision functions always count as parametric regardless of their control role, with only unlearned enforcement and sequencing logic in H) and then verify that the classifications in 4-7 are consistent under it.","section":"3 vs. 5.4 and 6.2","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The empirical basis of the taxonomy is the definition-guided corpus collection and per-study coding described in 2.4 (teacher-student configuration, knowledge form, updated components, mechanism, evaluation setting), but no coding instrument, inter-coder agreement measure, or inclusion/exclusion statistics are reported. Given that the survey's central organizing claim is that every core method admits a primary substrate under the 3 rule, the reproducibility of those assignments should be demonstrated rather than asserted. A short appendix containing the coding protocol, counts per substrate class, and one worked example for each of M, R, and H would make the taxonomy checkable and would also serve as the operationalization the field needs to apply the framework to new methods.","section":"2.4","authors_acknowledge":null,"authors_acknowledge_p":null}],"minor_comments":[{"comment":"The phrase 'preserving the repository's intended behavior intended behavior' contains a duplicated phrase that should be removed.","section":"2.2","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The word 'parameteric' appears in Fig. 2 and in 1.3 ('the parameteric model'); this is a typo for 'parametric.'","section":"Fig. 2 and 1.3","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The row label 'Ours2027' should be formatted consistently with the author-year style used for the other rows (e.g., 'Ours (2027)').","section":"Table 1","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"Given that Eq 22 defines the claim tuple (source, destination, evidence, transformation, Z_S) as the unit of evaluation, Table 3 would be more actionable if its final column also recorded the strongest supported source-to-destination arrow for each benchmark rather than only a general caveat; several rows (e.g., APIBank, ToolSandbox) currently leave the substrate attribution question implicit.","section":"Table 3","authors_acknowledge":null,"authors_acknowledge_p":null}],"recommendation":"major_revision"},"author_rebuttal":null,"desk_editor":{"model":"deepseek/deepseek-v4-flash","letter":"You should read this one. It's the first survey I've seen that squarely names the gap between model-centered distillation surveys (which stop at the student's parameters) and system-centered agent surveys (which don't organize around teacher-student transfer), then builds an actual framework for the joint space. The substrate taxonomy — parametric model, reusable artifacts, execution harness — is the right axis, and the 3x3 source-destination transfer matrix plus the hybrid criterion (Eq 21) is a real new synthesis, not recycled work. The evaluation section earns its keep too: separating acquisition from utility from fidelity from deployment, the gap-closure metric, and the marginal-contribution and synergy formulas (Eqs 27-28) are concrete and usable.\n\nThe authors are careful in the ways that matter. Scope is drawn honestly — in-context learning and transient consultation are out, self-evolution is in. They correctly refuse to relabel AFlow and Automated Design of Agentic Systems as distillation when no teacher role is identifiable, and they're upfront that black-box harness recovery yields an equivalence class, not unique structure.\n\nNow the soft spot, and it's the one worth refereeing. The classification rule in §3 says substrate membership is functional — \"how the retained knowledge participates in future task solving\" — and explicitly disavows physical format. But §6.2 assigns a learned router (RouteLLM) to M_S because the knowledge lives in its parameters, reserving H_S for the runtime that queries the router. The same decision function — which model to call — is parametric when learned and harness when written as rules. That's exactly the physical-format criterion §3 says doesn't decide. The paper sees the problem (§9.1 concedes different M/R/H combinations give the same behavior; §7.1 demands source-agnostic labels; §6.4 refuses a single label for joint changes), yet Sections 4-6 still hand out single-cell assignments for the corpus. The fix is cheap: an inter-coder reliability check on a sample of the corpus, or explicit boundary-case rulings per contested method. Without it, the taxonomy's organizing power rests on the authors' reading.\n\nMinor: the substrate vocabulary isn't wholly new — AgentDistill's MCP modules, memory distillation, and harness surveys already use pieces of it. The unified framing is the contribution, and the paper hedges appropriately; citing published results for derived claims is normal survey practice, not a flaw.\n\nWho it's for: anyone distilling agents, building agent memory/tool/workflow systems, or writing agent evaluation protocols. It deserves a serious referee; the classification issue is fixable in revision and doesn't undercut the core synthesis.","headline":"A genuinely useful organizing survey for agent-level knowledge transfer, with one real soft spot: the primary-substrate classification rule is not fully operational for learned control components.","tokens_in":43307,"tokens_out":2798,"duration_ms":21509},"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek/deepseek-v4-flash","headline":"This paper redefines knowledge distillation at the agent level, organizing the field by where transferred knowledge persists: in the model, in reusable artifacts, or in the execution harness.","keywords":["agent distillation","knowledge distillation","LLM-based agents","knowledge substrate","parametric knowledge","artifact knowledge","harness knowledge","cross-substrate transfer"],"falsifier":"Find one published method whose transferred knowledge resists single-substrate assignment under the paper's functional rule — for example, a teacher-generated workflow description that is stored as a retrievable artifact and simultaneously compiled into active runtime control, where removing it changes both retrieval and orchestration.","tokens_in":43625,"feed_emoji":"🤖","tokens_out":6245,"duration_ms":39554,"temperature":0.7,"pith_summary":"Modern agents are more than language models: they pair a parametric model with reusable artifacts (memories, tools, skills) and an execution harness that routes, verifies, and recovers. This paper argues that knowledge distillation, which traditionally asks a student model to imitate a teacher model, must be redefined at the level of the whole agent. It defines Agent Distillation as the persistent transfer of task-solving knowledge from a teacher agent into a student agent, and organizes the literature by where that knowledge is retained: inside the model, as external artifacts, or in the execution harness. On this substrate taxonomy, every agent-distillation method, and every cross-substrate or hybrid variant, can be classified and evaluated. The payoff is a common foundation for building, benchmarking, and safely maintaining agents that learn from experience and from other agents.","feed_headline":"Agent smarts live in model, artifacts, or harness","feed_subtitle":"A single map tells where transferred knowledge really lives — and how to test that it did.","key_machinery":"The load-bearing object is the agent realization $A=(M,R,H)$ — the parametric model $M$, reusable artifacts $R$, and execution harness $H$ — together with the knowledge-substrate rule that assignment to a category follows how retained knowledge participates in future task solving, not its file format or how it was produced. Three substrates organize the survey: parametric knowledge in learned computation (fine-tuned adapters, policies), artifact knowledge in addressable objects (memories, demonstrations, plans, code, tools), and harness knowledge in execution logic (routing, verification, recovery, orchestration). A transfer path is expressed as $K^x_T \\rightarrow E_T \\rightarrow K^y_S$ over a $3\\times 3$ source-destination matrix, and the hybrid criterion marks students where more than one substrate retains a teacher-derived change. A running example of a coding agent debugging an API regression carries the definition through teacher trace, filtered and abstracted knowledge, and persistent reuse on a fresh task.","core_discovery":"The paper's central move is to relocate distillation's unit of transfer from the model to the agent. It formalizes an agent as a triple $A=(M,R,H)$ of parametric model, reusable artifacts, and execution harness, and defines Agent Distillation as the persistent incorporation of teacher-derived knowledge $K_T$ into the student's components, yielding an updated realization $A^+_S$ that retains the knowledge after the teacher is gone. From this it derives a taxonomy: knowledge internalized in learned computation is parametric, knowledge externalized as addressable objects (memories, skills, tools) is artifact-based, and knowledge operationalized as execution and control logic is harness knowledge. Methods are classified by the primary substrate in which teacher knowledge persists, giving a $3\\times 3$ source-destination matrix for cross-substrate transfer and a hybrid criterion ($|Z_S|>1$) for endpoints where multiple substrates retain change. It also contributes an evaluation framework that separates transfer evidence from outcome and asks whether claimed retention is causally responsible for deployed utility.","pith_inferences":["A testable consequence of the taxonomy is that methods classed as parametric, artifact, or harness should show different failure and maintenance profiles — parameters are fastest but least auditable, artifacts are editable but impose retrieval cost, and harnesses guarantee ordering but break on interface change — so one could predict which substrate a given deployment constraint should favor.","The hybrid criterion suggests a design rule the paper only sketches: knowledge that is stable, frequent, and latency-sensitive should be internalized in the model, while explicit, exact, or rapidly changing knowledge belongs in artifacts, and conditional or safety-critical control belongs in the harness.","An external reader could turn the attribution framework into a predictive experiment: train the same student knowledge under two substrate allocations (parametric versus artifact-based) at equalized budgets, then measure not only utility but update cost, rollback ease, and portability across base models.","The definition implies a boundary case worth watching: black-box harness distillation recovers an observational equivalence class, not the teacher's unique organization, so porting third-party orchestration should be described as functional replication unless traced or causally identified."],"forward_implications":["Conventional model-level distillation, in-context prompting, and inference-time consultation are grouped as mechanisms or neighbouring work, not Agent Distillation, unless they produce a persistent change in the student agent.","Every existing agent-distillation method can be filed under one of three branches (parametric, artifact, harness), and cross-substrate or hybrid methods are characterized by both their transfer path and their deployment endpoint.","Evaluation should tie transfer evidence to causal contribution: reporting which student substrates changed and testing by removal or swap whether they matter, rather than reporting an end-task score alone.","Self-evolution counts as Agent Distillation when prior experience supplies teacher-side evidence that the same agent persistently retains for future episodes.","Benchmarks are organized by the deepest claim they support, with substrate attribution and teacher provenance required before a gain is credited to a particular component."],"supporting_citations":[{"why":"Establishes LLM-based agents as goal-directed systems capable of sustained task solving and environment interaction, grounding the agent-realization view the paper extends.","marker":"[80]"},{"why":"Canonical model-centered distillation survey whose framing the paper identifies as the gap to widen toward the agent level.","marker":"[24]"},{"why":"Foundational knowledge-distillation mechanism (soft outputs, feature alignment) that anchors the parametric-substrate branch.","marker":"[30]"},{"why":"Generalized Knowledge Distillation on student-generated sequences, the on-policy principle that extends into agentic parametric distillation.","marker":"[1]"},{"why":"Converts task feedback into verbal reflections stored in an episodic memory buffer, grounding the artifact branch's memory construction.","marker":"[84]"},{"why":"Maintains a growing library of executable code improved by environment feedback, grounding environment-grounded artifact externalization.","marker":"[99]"},{"why":"Induces commonly reused workflows from demonstrations and experience, illustrating the artifact-to-harness boundary in the taxonomy.","marker":"[104]"},{"why":"Represents agent collaboration as optimizable computational graphs, supplying the orchestration representation the harness branch builds on.","marker":"[145]"},{"why":"Packages teacher-generated knowledge into reusable protocol modules for training-free transfer to smaller students, exemplifying portable artifact distillation.","marker":"[77]"},{"why":"Infers harness behavior from black-box queries of a multi-agent system, the most direct instance of harness distillation discussed.","marker":"[15]"}],"fun_headline_variants":["Agent Distillation roadmap: model, artifacts, harness, and evaluation","Where agent knowledge lives: model, artifacts, harness, and beyond","Distilling agents across model, artifacts, and execution harness","Move distillation from model imitation to full agent transfer","Agent Distillation: persistent transfer across model, artifacts, harness"],"cache_read_input_tokens":0,"weakest_assumption_plain":"The load-bearing premise is that every agent realization decomposes exhaustively into model, artifacts, and harness, and that each distillation method can be assigned one primary substrate, so the taxonomy's categories do not overlap in practice.","fun_headline_variants_meta":{"raw":{"variants":["Agent Distillation roadmap: model, artifacts, harness, and evaluation","Where agent knowledge lives: model, artifacts, harness, and beyond","Distilling agents across model, artifacts, and execution harness","Move distillation from model imitation to full agent transfer","Agent Distillation: persistent transfer across model, artifacts, harness"]},"model":"deepseek/deepseek-v4-flash","effort":"low","cost_usd":0.000671,"raw_usage":{"cost":0.000670628,"is_byok":false,"cost_details":{"upstream_inference_cost":0.000670628,"upstream_inference_prompt_cost":9.4108e-05,"upstream_inference_completions_cost":0.00057652},"total_tokens":2936,"prompt_tokens":877,"completion_tokens":2059,"prompt_tokens_details":{"audio_tokens":0,"video_tokens":0,"cached_tokens":256,"cache_write_tokens":0},"completion_tokens_details":{"audio_tokens":0,"image_tokens":0,"reasoning_tokens":1974}},"tokens_in":877,"tokens_out":2059,"cost_source":"openrouter_usage","duration_ms":11064,"image_pages":[],"temperature":1.0,"billed_attempts":1,"reasoning_tokens":1974,"transport_provider":"openrouter","cache_read_input_tokens":0,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-09-30T23:24:36.084547+00:00","model_set":{"reader":"deepseek/deepseek-v4-flash"},"pipeline_version":"pith-pipeline@v1.4.1-daily-deepseek-or-medium","input_kind":null,"input_chars":null,"input_total_chars":null,"passes_missing":null,"published_standing":null,"published_confidence":null,"objection_audit":null,"run_record":null,"basis":{"kind":"not_recorded","source":null,"chars_given":null,"chars_total":null},"receipt":null,"falsifier":"Find one published method whose transferred knowledge resists single-substrate assignment under the paper's functional rule — for example, a teacher-generated workflow description that is stored as a retrievable artifact and simultaneously compiled into active runtime control, where removing it changes both retrieval and orchestration.","supporting_citations":[{"cited_title":"Toolformer: Language Models Can Teach Themselves to Use Tools , url =","cited_arxiv_id":null,"evidence_quote":"Establishes LLM-based agents as goal-directed systems capable of sustained task solving and environment interaction, grounding the agent-realization view the paper extends."},{"cited_title":"Knowledge Distillation: A Survey.International Journal of Computer Vision2021,129, 1789–1819","cited_arxiv_id":null,"evidence_quote":"Canonical model-centered distillation survey whose framing the paper identifies as the gap to widen toward the agent level."},{"cited_title":"Distilling the Knowledge in a Neural Network","cited_arxiv_id":"1503.02531","evidence_quote":"Foundational knowledge-distillation mechanism (soft outputs, feature alignment) that anchors the parametric-substrate branch."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Generalized Knowledge Distillation on student-generated sequences, the on-policy principle that extends into agentic parametric distillation."},{"cited_title":"Gopinath, Karthik Narasimhan, and Shunyu Yao","cited_arxiv_id":null,"evidence_quote":"Converts task feedback into verbal reflections stored in an episodic memory buffer, grounding the artifact branch's memory construction."},{"cited_title":"2024 , month = aug, publisher =","cited_arxiv_id":null,"evidence_quote":"Maintains a growing library of executable code improved by environment feedback, grounding environment-grounded artifact externalization."},{"cited_title":"TCOD: Exploring Temporal Curriculum in On-Policy Distillation for Multi-turn Autonomous Agents","cited_arxiv_id":"2604.24005","evidence_quote":"Induces commonly reused workflows from demonstrations and experience, illustrating the artifact-to-harness boundary in the taxonomy."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Represents agent collaboration as optimizable computational graphs, supplying the orchestration representation the harness branch builds on."},{"cited_title":"AgentDistill: Training-Free Agent Distillation with Generalizable MCP Boxes","cited_arxiv_id":"2506.14728","evidence_quote":"Packages teacher-generated knowledge into reusable protocol modules for training-free transfer to smaller students, exemplifying portable artifact distillation."},{"cited_title":"Agent Harness Distillation: Inference-Time Harness Extraction and Exploitation in Autonomous Multi-Agent Systems","cited_arxiv_id":"2607.28147","evidence_quote":"Infers harness behavior from black-box queries of a multi-agent system, the most direct instance of harness distillation discussed."}],"lane_accuracy":{"status":"not yet measured","lane":{"model":"deepseek/deepseek-v4-flash","input_kind":null},"url":"https://pith.science/accuracy"},"review_version":1}