Pith. sign in

REVIEW 3 major objections 5 minor 21 references

PAFFA: Premeditated Actions For Fast Agents

T0 review · 3 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read PAFFA turns a web task into one lightweight API call by precomputing browser interactions offline, cutting inference tokens by 87% while improving step accuracy over a per-step baseline.

desk verdict PAFFA has a plausible skill-caching architecture, but its headline accuracy claim is built on comparing PAFFA's relaxed human-relabeled scores against the baseline's exact-match scores, and a matched comparison does not support it. read the letter →

arxiv 2412.07958 v2 pith:SIGBRMGJ submitted 2024-12-10 cs.AI

Shambhavi Krishna , Zheng Chen , Yuan Ling , Xiaojiang Huang , Yingjie Li , Fan Yang , Xiang Li This is my paper · ORCID
classification cs.AI
keywords PAFFAActionLibraryLLMwebagentszero-shotpromptingtokenefficiencyMind2Webdynamicinterfacesprogramsynthesis
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper is trying to establish that an LLM's web-interaction reasoning can be done once, offline, and reused across many tasks, so that at runtime the model no longer has to read the page. Its method, PAFFA, uses a zero-shot LLM to generate parameterized browser-action functions, stores them in an Action Library, and at runtime asks the LLM only to match a request to the right function and fill in its arguments. On the Mind2Web benchmark the paper reports element accuracy of 0.74 versus 0.56 and step accuracy of 0.57 versus 0.50 against a per-step HTML-parsing baseline, together with an 87% reduction in runtime inference tokens. If correct, this would make autonomous web agents far cheaper to operate, more resilient to changing page structure, and adaptable to new sites without retraining.

What carries the argument

The Action Library is the central object: a persistent collection of parameterized Python functions, executed through Selenium, that encode verified click, type, and navigate patterns for clusters of similar tasks on a website. Dist-Map constructs entries by having the LLM extract task-agnostic interactive elements into a structured JSON view, verifying that view, and then generating scripts against the distilled elements. Unravel constructs entries by incrementally processing the full HTML of each page state, using the task goal and interaction history to generate code that can handle multiple actions per view. At runtime the LLM's only job is API retrieval and parameter extraction; the retrieved function drives the browser itself. Unravel doubles as the novelty handler: when retrieval fails or a page has changed, Unravel explores the new state and its successful trace is parameterized and inserted into the library, which is the mechanism that makes the approach adaptive without retraining.

What would settle it

Run PAFFA on a set of live pages after introducing realistic DOM changes, such as altered element IDs, moved buttons, or new login walls, and count how often the first retrieved action function completes the task without triggering Unravel; if a non-negligible share of requests falls back to Unravel, the reported 87% token reduction and the accuracy gap over the baseline shrink by that share.

Watch

Extended reading notes

Core claim

The central claim is that LLM reasoning sequences about web interaction generalize across prompts: the same distilled interaction pattern, stored as a parameterized API, can serve many tasks on the same website. PAFFA builds this library offline through two strategies — Dist-Map distills key interactive elements into a compact representation before writing scripts, while Unravel explores page by page with full HTML and generates interaction code with error handling. At runtime the model performs only lightweight intent recognition and slot filling, then executes the stored API directly, avoiding iterative HTML parsing. On an aggregate airlines-plus-shopping split, PAFFA reports 0.74 versus 0.56 element accuracy and 0.57 versus 0.50 step accuracy against the fine-tuned baseline, and roughly 25,000 tokens and one call per task versus roughly 197,000 tokens and 126 calls. Unravel also feeds successful traces from novel pages back into the library, giving the agent a training-free adaptation mechanism for unseen websites.

Load-bearing premise

The argument depends on the assumption that a library of reusable browser actions built offline from a small set of example pages stays usable on unseen tasks and on live pages as they change, so that a single lookup with no fresh view of the current page is enough for most requests.

Editorial extensions

If this is right

  • Web agents built this way run at roughly one lightweight LLM call per task, changing the operational cost of deploying assistants on real websites.
  • A training-free library that grows from successful explorations can absorb website changes over time, so the agent's performance does not depend on keeping the base model fine-tuned.
  • The cross-website results imply that zero-shot, precomputed interactions can transfer to new sites better than models that learn site-specific behavior from training splits.
  • If the generalization claim holds, other tasks on the same website need not repeat the expensive parsing and planning once a similar action is in the library.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The token accounting in the paper covers runtime only; the offline cost of library construction would dominate for sites with very few tasks, so the economic case is strongest when the library is amortized over repeated use.
  • A direct stress test the authors did not run: introduce adversarial page changes such as randomized element IDs, reordered layouts, or new overlays and count fallbacks to Unravel; high fallback rates would convert the token savings into robustness results.
  • Retrieval could be made safer by attaching a lightweight page fingerprint, such as the presence of key element IDs, so stale action functions are detected before execution, making the library self-maintaining.
  • Because generated action functions use concrete selectors, a natural next step is semantic locators that survive superficial redesigns; without that, the library may need frequent Unravel refreshes on fast-moving sites.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The paper introduces PAFFA, a web-agent method that precomputes an "Action Library" of parameterized browser-interaction scripts offline using a zero-shot LLM (Claude Sonnet 3.5), then at runtime performs lightweight API retrieval and parameter extraction instead of repeated HTML parsing. Two library-construction strategies are presented, Dist-Map and Unravel, with Unravel also serving as a runtime fallback for novel pages and as a mechanism for library evolution. The authors report evaluations on Mind2Web subsets, claiming robust accuracy (0.57 vs 0.50 step accuracy, 0.74 vs 0.56 element accuracy) and an 87% reduction in inference tokens, plus an LLM-based qualitative script-quality comparison. The core idea is clearly described, but the headline accuracy comparisons are undermined by a mismatch between the "Inexact" metric used for PAFFA and the "Exact" metric used for the baselines.

Significance. If the claimed efficiency and accuracy held, PAFFA would make a useful contribution: the idea of caching generalized interaction routines offline and executing them with a single lightweight call is a plausible way to reduce the inference cost of web agents, and the training-free construction is attractive. The paper is well structured and the architecture is described in enough detail to reproduce the main pipeline. However, the central accuracy claim is not supported by the evidence as presented, and the efficiency claim is conditional on the frequency of runtime fallback, which is not measured. The qualitative evaluation is self-assessment and does not provide independent evidence. The contribution may be salvageable with re-analysis and additional experiments, but the current version overstates its findings.

major comments (3)
  1. [§4.1, Tables 1-2, abstract, and §1] The headline accuracy comparisons are not on the same metric. The abstract's "0.57 vs. 0.50 step accuracy" and "element accuracy: 0.74 vs. 0.56" compare PAFFA's human-relabeled "Inexact" scores with MindAct's official "Exact" scores. On the Air.+All Shop. split, PAFFA's exact step accuracies are 0.34 (Dist-Map) and 0.32 (Unravel), below the T5-XL baseline's 0.500, and its exact element accuracies are 0.699 and 0.650 versus 0.562. No "Inexact" scores are reported for any baseline, and no annotation protocol or inter-annotator agreement is reported for the relabeling. Under the benchmark's official exact-match metric, PAFFA's step accuracy is not robust relative to the baseline; the central claim is therefore unsupported as stated.
  2. [§4.2, Table 3, and §3.3.3] The 87% token-reduction figure is computed "After Setup Usage Per Task/Request" and assumes one API-retrieval call with no Unravel fallback. The paper does not report how often API retrieval selects the wrong API or how often a previously reliable API encounters execution errors, even though §3.3.1 lists these as triggers for invoking Unravel. If such failures are frequent, the average runtime token count is higher than the happy-path estimate. The paper should report the empirical distribution of runtime calls and, if the claim concerns total inference cost, account for the offline library-construction and API-synthesis tokens; alternatively, the efficiency claim should be explicitly scoped to the post-setup happy path.
  3. [§4.3, Table 4] The qualitative script-quality scores are produced by Sonnet 3.5 judging scripts generated by Sonnet 3.5. The paper acknowledges that this is an emerging evaluation technique, but the scores are then used to conclude that Unravel produces higher-quality scripts and to reinforce its central role. Without a human-annotated validation set, a second judge, or inter-annotator agreement, these scores do not constitute independent evidence; at most they are a self-assessment and should be presented as such.
minor comments (5)
  1. [Section 2] The text cites "MindSearch (Ma et al., 2023)" but the reference list contains "LASER: LLM agent with state-space exploration for web navigation" by Kaixin Ma et al.; the citation-reference pair appears mismatched.
  2. [Tables 1 and 2] The captions do not define "Ex" and "Inex", nor do they explain that the PAFFA columns use zero-shot Sonnet 3.5 while the S3.5 column uses 3-shot prompting; please make this explicit.
  3. [Appendix C.1] The test sets are small (31, 8, and 17 tasks), so the reported differences across splits should be accompanied by confidence intervals or significance tests.
  4. [Section 4.2, Table 3] The per-task token estimate of 25,000 for PAFFA is not accompanied by the prompt template or tokenization details; please specify how the estimate was computed.
  5. [Throughout] The notation "Ex" and "Inexact" is used inconsistently (table headers versus prose); unify the terminology.

Circularity Check

2 steps flagged · score 5.0 of 10

Central accuracy claim mixes PAFFA-only human-relabeled Inexact scores with baselines' Exact scores, and qualitative script superiority rests on Sonnet 3.5 judging scripts generated by Sonnet 3.5.

  1. other [Section 4.1, Tables 1-2; Abstract]
    "Crucially, web task annotations can be ambiguous, and multiple valid interaction paths often exist. Therefore, in addition to exact match accuracy (‘Exact’), we perform human re-evaluation to account for functionally correct but non-annotated paths (‘Inexact’)."

    The headline step-accuracy claim '0.57 vs. 0.50' compares PAFFA's Inexact column on Air.+All Shop. (0.57 for both Dist-Map and Unravel) to MindAct T5-XL's Exact column (0.50). The Inexact score is a human relabeling of PAFFA's own non-matching outputs as functionally correct, with no reported human relabeling for the baseline. PAFFA's own Exact step accuracy on that split is 0.34/0.32, below the baseline's 0.50. Thus the claimed 'robust performance' is not a prediction from the Action Library, but an artifact of applying a more lenient, PAFFA-only scoring rule to the same outputs.

  2. other [Section 4.3; Appendix D; footnote 6]
    "We employ LLM-based evaluation using Sonnet 3.5 on the Airlines dataset subset. ... We employ Sonnet 3.5 for qualitative assessment, providing consistent evaluations at scale, though we acknowledge this is an emerging evaluation technique."

    The scripts being scored are generated by the same Sonnet 3.5 model that serves as the judge, using subjective rubrics (Script Task Alignment, Action Representation Fidelity, Script Efficiency). No independent human annotation or inter-annotator agreement is reported. The conclusion that 'Unravel consistently outperforms Dist-Map' and 'Unravel appears superior' therefore reduces to Sonnet 3.5's self-assessment of Sonnet 3.5 outputs, which the paper itself flags as 'an emerging evaluation technique.' This is a load-bearing step for the qualitative superiority claim, though not for the token-efficiency numbers.

full rationale

The paper does not rely on self-citations, imported uniqueness theorems, or a mathematical derivation whose output is identical to its input; the Action Library architecture is an empirical construction, and the token-count comparison is a straightforward accounting exercise, though it excludes offline library-construction tokens. However, two evaluation choices load the headline claims in a circular manner. First, the central accuracy comparison mixes PAFFA's human-relabeled Inexact scores with baselines' Exact scores, so the 'maintaining robust performance' claim is supported only by a more lenient, PAFFA-only scoring rule; PAFFA's Exact scores are below the baseline on the headline split. Second, the qualitative script-quality comparison is self-evaluation by the same model that generated the scripts, with no independent ground truth. The paper itself concedes the qualitative metric is 'an emerging evaluation technique' and that the Inexact human evaluation is resource-intensive. These are not cases of self-citation or ansatz smuggling, but they are instances where the claimed evidence reduces to the system's own relabeled or self-assessed outputs. Score 5 reflects partial circularity in the central performance narrative while acknowledging that the efficiency mechanism and exact-match numbers are not circular.

Assumptions & free parameters 0 free parameters · 4 assumptions · 1 invented entities

PAFFA rests on two major domain assumptions: that an LLM can generate reliable browser scripts zero-shot offline, and that a small library of parameterized APIs can cover unseen tasks and live sites without reading current HTML. The evaluation adds a third assumption that human-relabeled 'Inexact' accuracy and LLM self-scoring are valid evidence. No new physical entities are introduced; the Action Library is a software construct with no independent evidence beyond the paper's own experiments.

assumptions (4)
  • domain assumption Claude Sonnet 3.5 can generate correct, executable Selenium interaction scripts from zero-shot prompts on webpages it has never been fine-tuned for.
    All offline library construction relies on this; Section 3.2 and the evaluation in Section 4 would fail if scripts frequently contain errors. The paper reports no script-execution success rate.
  • domain assumption A small set of parameterized APIs per website covers unseen tasks and dynamic interface states.
    Runtime execution is a single API-retrieval call plus parameter extraction (Section 3.3.3), with no examination of the live page HTML. The paper does not measure retrieval failure rate or how often Unravel must be invoked.
  • domain assumption Mind2Web's static snapshots and human 'Inexact' relabeling are a valid measure of real-world task success.
    The headline accuracy uses the Inexact metric from an undescribed human re-evaluation (Section 4.1, Appendix C). The paper acknowledges the benchmark's lack of interactivity.
  • domain assumption LLM-based evaluation is a valid proxy for script quality.
    Section 4.3 uses Sonnet 3.5 to score scripts generated by Sonnet 3.5; the authors call this emerging but present scores as quantitative results.
invented entities (1)
  • Action Library
    purpose: A persistent store of parameterized Selenium APIs pre-computed by an LLM, used to replace runtime HTML parsing with a single API-retrieval call.
    The library is the central construct of PAFFA. Its effectiveness is only evidenced by the paper's own, partly self-referential evaluations; no external benchmark or independent reproduction is provided.

how reviews work

0 comments
Cite this review

Pith. "Pith review of PAFFA: Premeditated Actions For Fast Agents." pith.science (2026). https://pith.science/paper/SIGBRMGJ

@misc{pith2026241207958,
  author       = {Pith},
  title        = {Pith review of: PAFFA: Premeditated Actions For Fast Agents},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/SIGBRMGJ}},
  note         = {Machine review of arXiv:2412.07958}
}
read the original abstract

Modern AI assistants have made significant progress in natural language understanding and tool-use, with emerging efforts to interact with Web interfaces. However, current approaches that heavily rely on repeated LLM-driven HTML parsing are computationally expensive and error-prone, particularly when handling dynamic web interfaces and multi-step tasks. We introduce PAFFA (Premeditated Actions For Fast Agents), a method that makes LLMs faster and more accurate in completing tasks on the internet using a novel inference-time technique that requires no task-specific training. PAFFA constructs an 'Action Library', leveraging the parametric knowledge of the base LLM to pre-compute browser interaction patterns that generalize across tasks. By strategically re-using LLM inference across tasks - either via 'Dist-Map' for task-agnostic identification of key interactive web elements, or 'Unravel' for first-encounter, stateful exploration of novel tasks/sites) - PAFFA drastically reduces inference time tokens by 87% while maintaining robust performance (achieving 0.57 vs. 0.50 step accuracy compared to baseline). Further, Unravel's ability to update its action library based on explorations allows generalization and adaptation to unseen websites. In sum, this work exhibits that LLM reasoning sequences can generalize across prompts, offering a way to scale inference-time techniques for internet-scale data with sublinear token count.

Figures

Figures reproduced from arXiv: 2412.07958 by the authors.

Figure 2
Figure 2. Grouping tasks solvable by one API [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Creating APIs per group. The Action Library is constructed by leveraging the inherent document understanding, reasoning, and code generation capabilities of LLMs 2 . 3.2 Initial Library Construction Strategies 3.2.1 Strategy 1: Dist-Map (Distillation then Mapping) The goal of this method is to search for reusable actions that will generalize across interface variations. We hypothesize that leveraging the LLM’s inher… view at source ↗
Figure 4
Figure 4. Common workflow of existing solutions like MindAct ( [PITH_FULL_IMAGE:figures/full_fig_p011_4.png] view at source ↗
Figures from the paper (1 more)
Figure 5
Figure 5. Figure 5: Using Action Library. C Dataset We evaluate our algorithm on the Mind2Web benchmark (Deng et al., 2023), which provides annotated actions for diverse tasks across real-world websites. This dataset’s key advantage is its use of complex, real-world websites rather than s…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

21 extracted references · 14 canonical work pages

  1. [1]

    URL https://www.selenium.dev/documentation/webdriver/

    Webdriver. URL https://www.selenium.dev/documentation/webdriver/. Accessed: 2024-12-02

    work page 2024

  2. [2]

    Claude 3.5 sonnet

    Anthropic. Claude 3.5 sonnet. https://www.anthropic.com, 2025. Large language model developed by Anthropic

    work page 2025

  3. [3]

    Zhao, Yanping Huang, Andrew M

    Hyung Won Chung, Le Hou, Shayne Longpre, Barret Zoph, Yi Tay, William Fedus, Yunxuan Li, Xuezhi Wang, Mostafa Dehghani, Siddhartha Brahma, Albert Webson, Shixiang Shane Gu, Zhuyun Dai, Mirac Suzgun, Xinyun Chen, Aakanksha Chowdhery, Alex Castro - Ros, Marie Pellat, Kevin Robinson, Dasha Valter, Sharan Narang, Gaurav Mishra, Adams Yu, Vincent Y. Zhao, Yanp...

    work page 2024

  4. [4]

    Mind2web: Towards a generalist agent for the web

    Xiang Deng, Yu Gu, Boyuan Zheng, Shijie Chen, Samual Stevens, Boshi Wang, Huan Sun, and Yu Su. Mind2web: Towards a generalist agent for the web. In NeurIPS, 2023

    work page 2023

  5. [5]

    Huang, Mustafa Safdari, Yutaka Matsuo, Douglas Eck, and Aleksandra Faust

    Izzeddin Gur, Hiroki Furuta, Austin V. Huang, Mustafa Safdari, Yutaka Matsuo, Douglas Eck, and Aleksandra Faust. A real-world webagent with planning, long context understanding, and program synthesis. In ICLR . OpenReview.net, 2024

    work page 2024

  6. [6]

    Webvoyager: Building an end-to-end web agent with large multimodal models

    Hongliang He, Wenlin Yao, Kaixin Ma, Wenhao Yu, Yong Dai, Hongming Zhang, Zhenzhong Lan, and Dong Yu. Webvoyager: Building an end-to-end web agent with large multimodal models. In ACL (1) , pp.\ 6864--6890. Association for Computational Linguistics, 2024

    work page 2024

  7. [7]

    Deberta: decoding-enhanced bert with disentangled attention

    Pengcheng He, Xiaodong Liu, Jianfeng Gao, and Weizhu Chen. Deberta: decoding-enhanced bert with disentangled attention. In ICLR . OpenReview.net, 2021

    work page 2021

  8. [8]

    Tptu-v2: Boosting task planning and tool usage of large language model-based agents in real-world industry systems

    Yilun Kong, Jingqing Ruan, Yihong Chen, Bin Zhang, Tianpeng Bao, Shiwei Shi, Du Qing, Xiaoru Hu, Hangyu Mao, Ziyue Li, Xingyu Zeng, Rui Zhao, and Xueqian Wang. Tptu-v2: Boosting task planning and tool usage of large language model-based agents in real-world industry systems. In EMNLP (Industry Track) , pp.\ 371--385. Association for Computational Linguist...

    work page 2024

Show all 21 references
  1. [9]

    Markuplm: Pre-training of text and markup language for visually rich document understanding

    Junlong Li, Yiheng Xu, Lei Cui, and Furu Wei. Markuplm: Pre-training of text and markup language for visually rich document understanding. In ACL (1) , pp.\ 6078--6087. Association for Computational Linguistics, 2022

    2022

  2. [10]

    Reinforcement learning on web interfaces using workflow-guided exploration

    Evan Zheran Liu, Kelvin Guu, Panupong Pasupat, Tianlin Shi, and Percy Liang. Reinforcement learning on web interfaces using workflow-guided exploration. In ICLR (Poster) . OpenReview.net, 2018

    2018

  3. [11]

    Weblinx: Real-world website navigation with multi-turn dialogue

    Xing Han Lu, Zdenek Kasner, and Siva Reddy. Weblinx: Real-world website navigation with multi-turn dialogue. In ICML . OpenReview.net, 2024

    2024

  4. [12]

    LASER: LLM agent with state-space exploration for web navigation

    Kaixin Ma, Hongming Zhang, Hongwei Wang, Xiaoman Pan, and Dong Yu. LASER: LLM agent with state-space exploration for web navigation. CoRR, abs/2309.08172, 2023

    2023 arXiv

  5. [13]

    FLIN: A flexible natural language interface for web navigation

    Sahisnu Mazumder and Oriana Riva. FLIN: A flexible natural language interface for web navigation. In NAACL-HLT , pp.\ 2777--2788. Association for Computational Linguistics, 2021

    2021

  6. [14]

    Autonomous evaluation and refinement of digital agents

    Jiayi Pan, Yichi Zhang, Nicholas Tomlin, Yifei Zhou, Sergey Levine, and Alane Suhr. Autonomous evaluation and refinement of digital agents. CoRR, abs/2404.06474, 2024 a

    2024 arXiv

  7. [15]

    Webcanvas: Benchmarking web agents in online environments

    Yichen Pan, Dehan Kong, Sida Zhou, Cheng Cui, Yifei Leng, Bing Jiang, Hangyu Liu, Yanyi Shang, Shuyan Zhou, Tongshuang Wu, and Zhengyang Wu. Webcanvas: Benchmarking web agents in online environments. CoRR, abs/2406.12373, 2024 b

    2024 arXiv

  8. [16]

    From pixels to ui actions: Learning to follow instructions via graphical user interfaces

    Peter Shaw, Mandar Joshi, James Cohan, Jonathan Berant, Panupong Pasupat, Hexiang Hu, Urvashi Khandelwal, Kenton Lee, and Kristina N Toutanova. From pixels to ui actions: Learning to follow instructions via graphical user interfaces. Advances in Neural Information Processing S...

    2023

  9. [17]

    Synapse: Leveraging few-shot exemplars for human-level computer control

    Longtao Zheng, Rundong Wang, and Bo An. Synapse: Leveraging few-shot exemplars for human-level computer control. CoRR, abs/2306.07863, 2023

    2023 arXiv

  10. [18]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

  11. [19]

    @esa (Ref

    \@ifxundefined[1] #1\@undefined \@firstoftwo \@secondoftwo \@ifnum[1] #1 \@firstoftwo \@secondoftwo \@ifx[1] #1 \@firstoftwo \@secondoftwo [2] @ #1 \@temptokena #2 #1 @ \@temptokena \@ifclassloaded agu2001 natbib The agu2001 class already includes natbib coding, so you should ...

  12. [20]

    \@lbibitem[] @bibitem@first@sw\@secondoftwo \@lbibitem[#1]#2 \@extra@b@citeb \@ifundefined br@#2\@extra@b@citeb \@namedef br@#2 \@nameuse br@#2\@extra@b@citeb \@ifundefined b@#2\@extra@b@citeb @num @parse #2 @tmp #1 NAT@b@open@#2 NAT@b@shut@#2 \@ifnum @merge>\@ne @bibitem@firs...

  13. [21]

    @open @close @open @close and [1] URL: #1 \@ifundefined chapter * \@mkboth \@ifxundefined @sectionbib * \@mkboth * \@mkboth\@gobbletwo \@ifclassloaded amsart * \@ifclassloaded amsbook * \@ifxundefined @heading @heading NAT@ctr thebibliography [1] @ \@biblabel @NAT@ctr \@bibset...

    2025

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.