Pith.

Computer Vision and Pattern Recognition · arXiv:2611.01860 v1 · submitted 1 Nov 2026 · 291 references

Advances in Agentic Computer Graphics

Hengyu Liu, Jihyeon Je, Ruiqi Wang, Mingxuan Wu, Haven Feng, Qixing Huang, Georgios Pavlakos, Hao Zhang

show 6 more authors

Jiajun Wu, Maneesh Agrawala, Leonidas Guibas, R. Kenny Jones, Shangzhe Wu, Guandao Yang

The paper says

“Computer graphics is built on code. Large language model (LLM) agents can now generate graphics code and iteratively refine it through various feedback.”

more of the abstract

This capability opens the door to a potentially new paradigm of computer graphics, with graphics algorithms increasingly built around agentic systems to automate progressively complex tasks. Over the past years, research on agentic graphics has expanded rapidly across a diverse range of tasks. This rapid growth makes it increasingly difficult for manual literature reviews to keep pace with new developments. At the same time, similar ideas are often described using different terminology across these tasks, further complicating efforts to develop a systematic understanding of the field. In this report, we address these challenges through two contributions. First, we introduce a unified taxonomy for characterizing agentic systems across diverse computer graphics tasks. Second, we develop an agentic literature system that can assist human researchers in analyzing existing and newly published papers at scale. Building on these contributions, we analyze emerging research trends and discuss open challenges in coordinating agents and tools across graphics tasks, designing interfaces and representations for human-agent collaboration, sustaining long-horizon planning and refinement, and reliably evaluating interactive graphics systems.

from the abstract

Pith says

Machine review · deepseek-v4-flash · 4 Nov 2026 · what was read was not recorded · not checked by a person

It can be wrong. Challenge this review or reply to it.

The passes disagreed. Reader: holds up. Skeptic: holds up only with conditions. Referee: major revision.

A well-organized taxonomy for agentic graphics, but the trend analysis relies on unverified automatic labels and should be treated as suggestive, not conclusive. This line is wrong

3 major objections · 6 minor

How often objections from reviews like this one are right: not yet measured

Are you an author? What you can do.

Not checked: figures, code, what the 291 references say.

The paper says, at Sec. 4.2.2, Fig. 5

The review did not quote this passage.

A classifier read this sentence of the paper as acknowledging the concern (probability 0.77):

“We note that while these checks identify missing or inconsistent entries, they do not verify the correctness of the classifications.”

Pith says: Objection 1 of 3 · major

The quantitative trend analyses (e.g., 'local patching rising from 50.0% to 83.0%') are derived entirely from taxonomy labels assigned by the automated literature survey system. The paper explicitly states that curation checks 'do not verify the correctness of the classifications' (sentence b238).

more of objection 1

Manual verification was limited to paper relevance (Sec. 4.2.1), not to taxonomy labels. Without any validation of label accuracy (e.g., human inter-annotator agreement, per-category precision/recall), the reported percentages may reflect classification noise rather than genuine trends. This is a load-bearing issue because the trend analysis is a central empirical contribution of the paper. The authors should either (a) provide a validation study on a representative sample, or (b) clearly reframe these figures as exploratory and remove the quantitative precision (e.g., replace exact percentages with qualitative ranges).

The paper says, at Sec. 3, overall taxonomy

The review did not quote this passage.

Pith says: Objection 2 of 3 · major

The taxonomy is treated as a complete design space covering all agentic graphics systems. While the paper notes in Sec. 10 that categories can be expanded, the analysis in Sec. 4.2.2 relies on the current fixed set of dimensions and options.

more of objection 2

The paper would benefit from a more explicit discussion of potential blind spots—e.g., whether the four dimensions adequately capture systems that involve mixed-initiative human–agent collaboration or that use non-LLM decision-making components alongside LLMs. This could be addressed by adding a limitations paragraph to Sec. 3.

The paper says, at Sec. 4.1

The review did not quote this passage.

Pith says: Objection 3 of 3 · major

The agentic literature survey system is described at a high level, but key details needed for reproducibility are omitted: the exact prompt template for taxonomy annotation, the model configuration (e.g., temperature, number of reasoning steps), the tools used for paper reading and evidence extraction, and the criteria for the 'insufficient supporting evidence' flag.

more of objection 3

The paper states that the system will be released publicly, but until then the annotation process is not verifiable. The authors should provide a more detailed technical appendix or release the system as part of the supplementary material.

6 minor objections

The paper says, at Sec. 2, Fig. 2b

Pith says: minor objection 1 of 6

The font in Fig. 2b is very small; consider enlarging it or using a legend that lists the terms in a table for readability.

The paper says, at Sec. 3.4.1

Pith says: minor objection 2 of 6

The sentence 'These types differ in how much of an operation the agent must specify and how much is handled by the execution environment' is a helpful summary, but it could be made more precise by adding a short example contrasting code/script vs. tool/function call.

show 4 more

The paper says, at Tables 2, 3, 4

Pith says: minor objection 3 of 6

The tables are dense and use many icons. Consider adding a one-sentence caption that guides the reader to the most important pattern in each table (e.g., 'most methods use off-the-shelf VLMs with local patch updates').

The paper says, at Sec. 4.2.2, Fig. 5

Pith says: minor objection 4 of 6

The subfigures (a.1, a.2, a.3) are labeled only in the caption; label them on the figure itself for easier reference.

The paper says, at Sec. 9

Pith says: minor objection 5 of 6

The discussion of 'creativity and attribution' is welcome but brief. Consider adding a reference to recent work on human–AI co-creation to strengthen the context.

The paper says, at Sec. 2.1

Pith says: minor objection 6 of 6

The phrase 'these generators fall outside our scope' is clear, but it could be more explicit about why: they do not use an LLM to decide actions, as stated earlier. Repetition here is fine, but the justification should be consistent.

What would settle it

Randomly sample 50 papers from the corpus; have two domain experts independently assign all relevant taxonomy labels (dimensions in Sec 3). Compute per-category Cohen's kappa between experts and between each expert and the automated system. Report precision and recall for the automated labels using the majority expert vote as ground truth. If mean per-category agreement (expert vs automated) falls below 75% for any category appearing in more than 10% of the sample, the trend claims in Sec 4.2.2 should be caveated as preliminary and the paper should release the validation data.

from the skeptic's reading

How the review went

4 readings by deepseek-v4-flash · 4 Nov 2026

  1. Reader holds up confidence high
  2. Skeptic partly agrees: says it holds up only with conditions 24 s · 49,026 tokens read, 2,108 written Trend analyses in Sec 4.2.2 rely on automated taxonomy labels whose correctness is not verified; the reported percentages may be artifacts of classification noise.
  3. Editor wrote the verdict above 22 s · 49,851 tokens read, 1,695 written
  4. Referee major revision: 3 major objections, 6 minor 26 s · 49,401 tokens read, 2,275 written

Not run on this paper: the simulated authors' defence.

Pith did not check this review's objections against the paper.

The review in full

The review in brief, machine-written

Hi colleague,

Here's my take on the agentic graphics survey (2611.01860).

The paper's main contribution is a clean four-dimensional taxonomy—specification, base model, harness, graphics interface and tooling—that makes it possible to compare systems that use different terminology. The survey covers a huge range of recent work (350 papers) and the application sections (2D, static 3D, dynamic 3D) are thorough. The open problems discussion is thoughtful. The paper is clearly written.

What's good: the taxonomy is the lasting contribution. The authors also built an automated literature survey system that classifies papers using their taxonomy. As a way to keep up with a fast-growing field, the idea is sound. The manual verification of paper relevance is a step toward quality control.

What's soft: the trend analysis in Section 4.2.2—the main empirical claim—rests entirely on taxonomy labels assigned by the automated system. The paper explicitly says (sentence b238) that the curation checks "do not verify the correctness of the classifications." Manual verification was limited to relevance, not the labels themselves. So numbers like local patching rising from 50% to 83% could reflect classification noise. The stress-test note on this is accurate. Without inter-annotator agreement or per-category precision/recall, these percentages are suggestive but not robust.

That said, the taxonomy and qualitative comparisons survive this issue. The paper acknowledges the limitation. If the authors want the trend analysis to carry weight, they need to validate the labels. For now, I'd treat those numbers as rough indicators.

Who this is for: anyone entering agentic graphics, or anyone who wants a structured map of the field. The taxonomy is worth citing. I'd accept peer review, but would ask for either label validation or a significant softening of the quantitative claims.

Best, [Your Name]

The strongest objection

The paper's strongest empirical contribution—trend analyses showing, e.g., local patching rising from 50.0% to 83.0% (Fig 5b.1)—is derived entirely from taxonomy labels assigned by the automated literature survey system. The paper explicitly states that the system's curation checks "do not verify the correctness of the classifications" (sentence b238). Manual verification was limited to paper relevance (Sec 4.2.1), not to the taxonomy labels themselves. Without ground-truth validation, classification errors of unknown magnitude propagate into every quantitative claim in Sec 4.2.2. For example, if the system systematically mislabels 'local patch' vs 'regeneration' for a subset of papers, the observed temporal trend could be spurious. The paper's central assertion—that the taxonomy enables systematic cross-task comparison and trend discovery—depends on the reliability of these automatically generated numbers. The absence of any validation protocol (e.g., human inter-annotator agreement, per-category precision/recall) is a significant gap that weakens the empirical support for this claim.

The reader identified the reliability of automated classifications as a weakest assumption. I agree that this is a critical vulnerability and sharpen it to a specific, testable deficiency: the quantitative trend analyses in Sec 4.2.2 are unvalidated. The taxonomy itself is a valuable conceptual contribution, and the automated system is presented as assistive, so the paper does not fail entirely. However, the paper makes explicit empirical claims (e.g., percentages and trends) that are not supported by evidence of classification accuracy. A conditional accept that requires validation of a random sample (or explicit softening of the claims) would strengthen the paper's empirical grounding without rejecting its main organizational contribution.

The referee's report

The taxonomy provides a well-structured vocabulary that can unify descriptions across the fragmented agentic graphics literature. The automated survey system is a novel methodological contribution that, if made reliable, could lower the barrier to keeping surveys current. The qualitative synthesis of design patterns (Secs. 6–8) is valuable for researchers entering the field. If the quantitative trend claims can be validated, the paper would also offer a data-driven view of the field's evolution.

Pith's reading of the paper (a machine's summary, not the authors')

This survey paper argues that research on LLM-based agentic graphics systems—where a language model writes code, runs it in a graphics environment, and revises based on feedback—has grown so fast and with such inconsistent terminology that manual understanding is no longer sufficient. The authors propose a unified taxonomy that classifies any such system along four dimensions: specification, base model, harness, and graphics interface and tooling. To demonstrate and deploy this taxonomy, they build an agentic literature system that automatically discovers relevant papers, reads them, assigns taxonomy labels, and curates structured reports. Using this system on 350 papers, they identify several design patterns, such as the dominance of visual feedback (77 % of systems) and a shift toward local patch updates (rising to 83 % over the analyzed period). The paper closes with open challenges around multi-agent coordination, long-horizon planning, human–agent alignment, and reliable evaluation.

What carries the argument. The four-dimensional taxonomy (specification, base model, harness, graphics interface and tooling) is the central organizing framework; it provides a common vocabulary for describing and comparing agentic graphics systems across inconsistent terminologies. The agentic literature survey system, built on top of a SWE-agent-style harness with PaperQA2-style evidence extraction, operationalizes the taxonomy by reading full papers, assigning labels, and curating structured records.

Core claim. The paper establishes that the design space of agentic graphics systems can be captured by four axes—specification (task, input, output), base model (type and training), harness (planning, feedback, memory), and graphics interface and tooling (action type, software, representation, tools, resources)—each with concrete sub-options. Using this taxonomy, the authors find that most systems rely on off-the-shelf base models, harness designs are highly diverse, code and structured specifications dominate action types, and feedback patterns vary strongly by task domain (e.g., CAD favors full regeneration while scene generation favors local patches). The automated survey system itself is a demonstration that large-scale, taxonomy-driven literature curation can be performed by an agent, though the authors note that classification correctness is not systematically verified.

The premise it rests on. The four dimensions and their subcategories capture the essential design choices of all agentic graphics systems sufficiently to enable meaningful cross-task comparison; the paper acknowledges this is a design choice that may need expansion as new approaches appear.

If the paper is right (Pith's inferences, not the authors')

Researchers can use the taxonomy to position their own system design explicitly, enabling apples-to-apples comparisons with prior work even when different application terms are used.

The observed shift toward local patch updates suggests that future system designs should prioritize targeted edits over full regeneration to improve efficiency and preserve earlier progress.

The dominance of visual feedback points to an increasing reliance on VLMs as critics; future work may need to address VLM blind spots in geometric and physical reasoning.

The automated survey system, if publicly released as planned, could become a living bibliography that keeps pace with the field's rapid growth, reducing the lag of manual reviews.

The taxonomy's four dimensions are a design choice, not a theorem; a different partitioning (e.g., separating feedback from planning more sharply) might prove equally useful, and the paper invites such refinements as new approaches emerge.

The automated system's classification accuracy is unverified; a human validation study comparing its labels against expert annotations would be a natural next step before relying on trend numbers.

If the taxonomy is adopted widely, it could enable meta-analyses that cross-task comparisons of architecture choices (e.g., which harness patterns correlate with success on dynamic vs. static tasks) become routine.

The open challenge of evaluating interactive graphics systems is particularly acute; the paper notes stochastic runs and proprietary models make disentangling system design from model capability difficult, implying that a shared evaluation protocol would greatly accelerate progress.

Check this review

Pith stored no record of this review's passes and took no fingerprint of it. Reviews made by code from 13 Nov 2026 on carry both.

How often reviews like this one are right

Reviews like this one: written by deepseek-v4-flash, where what was read was not recorded.

How often objections from reviews like this one are right: not yet measured

Every value above, in the review's public record (JSON)

Reviews by people

This review was written by a machine. No person has reviewed this paper on Pith yet. A scientist listed on Pith can review it for a fee, and their review is published here whatever it concludes.

Request a human review

How Pith checks papers · The full audit of this paper

The paper

Abstract

Computer graphics is built on code. Large language model (LLM) agents can now generate graphics code and iteratively refine it through various feedback. This capability opens the door to a potentially new paradigm of computer graphics, with graphics algorithms increasingly built around agentic systems to automate progressively complex tasks. Over the past years, research on agentic graphics has expanded rapidly across a diverse range of tasks. This rapid growth makes it increasingly difficult for manual literature reviews to keep pace with new developments. At the same time, similar ideas are often described using different terminology across these tasks, further complicating efforts to develop a systematic understanding of the field. In this report, we address these challenges through two contributions. First, we introduce a unified taxonomy for characterizing agentic systems across diverse computer graphics tasks. Second, we develop an agentic literature system that can assist human researchers in analyzing existing and newly published papers at scale. Building on these contributions, we analyze emerging research trends and discuss open challenges in coordinating agents and tools across graphics tasks, designing interfaces and representations for human-agent collaboration, sustaining long-horizon planning and refinement, and reliably evaluating interactive graphics systems.

The paper's own sentences, arranged, kept beside the review and separate from it. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. No sentence in this record was written by a model.

abstract compiled · 2026-11-03

The authors' main claim, as the classifier picked it out 0.63

“This capability opens the door to a potentially new paradigm of computer graphics, with graphics algorithms increasingly built around agentic systems to automate progressively complex tasks.”

How the main claim is stated: hedged possibility 0.96

What the abstract itself states, as probabilities: claims first 0.06 · states limitation 0.13 · human participants 0.13 · controlled comparison 0.02 · randomized assignment 0.01 · states quantitative result 0.03 · compares with prior results 0.06 · mentions code or data availability 0.03

The abstract, sentence by sentence

This table labels the abstract only (arXiv listing abstract v1). The number is the probability the classifier gave the label.

SentenceRoleThe authors' own workStates the main conclusionStrength
Computer graphics is built on code. background 1.00 0.04 0.02 asserted 1.00
Large language model (LLM) agents can now generate graphics code and iteratively refine it through various feedback. background 1.00 0.08 0.05 asserted 0.92
This capability opens the door to a potentially new paradigm of computer graphics, with graphics algorithms increasingly built around agentic systems to automate progressively complex tasks. background 0.78 0.07 0.16 hedged possibility 0.96
Over the past years, research on agentic graphics has expanded rapidly across a diverse range of tasks. background 0.99 0.07 0.03 asserted 1.00
This rapid growth makes it increasingly difficult for manual literature reviews to keep pace with new developments. background 0.72 0.15 0.02 asserted 0.97
At the same time, similar ideas are often described using different terminology across these tasks, further complicating efforts to develop a systematic understanding of the field. background 0.61 0.11 0.02 asserted 1.00
In this report, we address these challenges through two contributions. contribution 0.98 0.58 0.05 asserted 0.98
First, we introduce a unified taxonomy for characterizing agentic systems across diverse computer graphics tasks. contribution 1.00 0.96 0.08 asserted 1.00
Second, we develop an agentic literature system that can assist human researchers in analyzing existing and newly published papers at scale. contribution 1.00 0.96 0.08 asserted 1.00
Building on these contributions, we analyze emerging research trends and discuss open challenges in coordinating agents and tools across graphics tasks, designing interfaces and representations for human-agent collaboration, sustaining long-horizon planning and refinement, and reliably evaluating interactive graphics systems. contribution 0.78 0.78 0.36 asserted 0.94

What kind of paper the classifier reports this is: method or tool 0.57 · evidence from literature 0.97

What was examined, and what was not

Sources read

  • arXiv listing abstract v1

Not examined

  • full text
  • figures and tables
  • references and citing sentences
  • relations to other papers

Levels run

  • P abstract

Not run

  • S sections
  • T body sentences
  • C citations
  • R cross-paper relations

How the labels were made

Compiled by classifier-1.13.0 via TypeSafe API, question bank abstract-p1-20260923, 2026-11-03: 1 request, 7,354 input tokens.

Every quoted sentence on this record is the authors'. Every label is from Pith's closed vocabulary, with the probability shown beside it. This record as JSON

Figures (19)
Figure 1 of the paper
Figure 1. Agentic Systems for Computer Graphics use language models to operate graphics tools and refine their outputs through feedback. From left to right: drawing sketches through sequential strokes [VSZ*25] and refining SVG illustrations from rendered feedback [WYZ*26]; editing 3D scenes from language instructions [SWJ*26] and modifying geometry and appearance to match target images [GHJ*25]; gener￾ating articulated 3D assets [ZLL*26b] and inferring executable scene programs from videos [SKK*26]. Abstract Computer graphics is built on code. Large language model (LLM) agents can now…
Figure 2 of the paper
Figure 2. Motivation for organizing the agentic graphics litera￾ture. (a) Rapid growth of our collected paper corpus. (b) Different names used across papers for feedback modules (top; Sec. 3.3.2) and planning modules (bottom; Sec. 3.3.1), motivating a unified taxonomy for comparing system designs. diate graphics results or the execution of those operations is used to select or revise subsequent actions during the task. This definition is purposefully inclusive: we impose no restric￾tions on the graphics representation, software platform, training strategy, or level of automation. These…
Figure 3 of the paper
Figure 3. Design Space of Agentic Graphics Systems. We taxonomize our design space into four dimensions: specification, base model, harness, graphics interface and tooling. We detail subclassification within each system following the taxonomy. • ó An existing 3D representation supplies content for modifica￾tion, such as Blender scenes with procedural geometry, material graphs, and lighting configurations [HYG24], or monolithic 3D Gaussian Splatting scenes that can be decomposed into individ￾ually editable objects [HGC*26]. • A A document provides source content for graphics authoring, such as…
Figure 4 of the paper
Figure 4. Overview of our agentic literature survey system. The agent searches for candidate papers and organizes them into a reading queue, then examines text and figures to assess scope and assign taxonomy labels (Sec. 3). Curation tools check report structure and save annotations with supporting evidence, design insights, and limitations. These records provide context for subsequent discovery and reading. defining articulated parts [ZLL*26b], CAD construction graphs recording modeling operations [LNC*26], and domain-specific programs expressing component relationships [WWH*26]. Retain￾ing…
Figure 5 of the paper
Figure 5. Design patterns in the annotated corpus. Analysis in (a) observation types, (b) update strategies, (c) resources. (a.1–a.3) Task distributions by observation category; (b.1) Temporal trends in update strategy and (b.2) Update strategy by task; (c.1) Temporal trends in resources and (c.2) Resource counts by task. Task percentages use all assignments within each observation category; temporal percentages use records annotated for each field and period. 4.2.2. Search Analysis Our agentic literature survey system enables us to systematically analyze a large body of existing works using…
Figure 6 of the paper
Figure 6. Existing benchmarks and datasets for evaluating static and dynamic artifacts. We show examples of datasets for CAD [MKC*25], Blender objects [GSY*26], and scene generation [LLL*25] for static, and articulated objects [ZLL*26b], dynamic scenes [TSM*26], and motion graphics [MA25] generation for dynamic outputs. form graphics tasks is becoming increasingly important. General￾purpose vision-language and spatial understanding benchmarks provide an important foundation for this evaluation, measuring ca￾pabilities such as object recognition, spatial relationships, visual reasoning, and…
Figure 9 of the paper
Figure 9. LL3M [LCD*25] generates editable Blender programs using retrieved API documentation, then uses visual feedback to guide iterative code revisions and verify whether they resolve the identified issues. In this example, an initially missing hydrant cap is added but remains detached, prompting a further revision to lower it into place before a final visual check confirms its attachment. geometric relationships constrain the resulting figure [ZWY*26]. Others focus on reconstructing existing visual content into editable representations. DeTikZify [BPE24] targets the translation of…
Figure 10 of the paper
Figure 10. Left: BlenderAlchemy [HYG24] uses visual feedback to revise material programs, adjusting shader parameters and graph structure to match text instructions. Right: ShapeCraft [ZJLD25] uses the object’s component structure to guide texture generation, aligning surface appearance with the text description
Figure 11 of the paper
Figure 11. ViSculpt [PPZ*26] decomposes language requests into localized edits on existing meshes using a compact set of sculpting primitives in Blender’s GUI. View selection and target masks con￾strain where these actions operate, while before-and-after visual feedback guides refinement of localization and brush parameters. Successful operations are stored in a reference library for reuse. make slide generation more controllable and adaptable after ini￾tial creation. MemSlides [JXZY26] uses hierarchical memory to support personalization and iterative revision, SeaSlides [FWZ26] introduces a…
Figure 12 of the paper
Figure 12. Top: CADCodeVerify [ATZ*25] converts design re￾quirements into visual questions and uses the answers to guide code refinement, addressing discrepancies between requested and gener￾ated shapes. Bottom: Seek-CAD [LLS*26] compares intermediate shapes with the corresponding design reasoning to guide code re￾finement, exposing geometry that may be obscured in final views. verification, so that each requirement can be assessed in a state that exposes the relevant geometry. For reference-based reconstruction, CADReasoner [KKV*26] reconstructs CAD programs from target surface geometry,…
Figure 13 of the paper
Figure 13. CAD-Assistant [MKC*25] uses a VLM planner to ex￾ecute text- and sketch-guided CAD operations in FreeCAD. Ren￾dered views and structured geometry summaries enable the plan￾ner to adjust subsequent edits as the design evolves
Figure 16 of the paper
Figure 16. FirePlace [HBT*25] grounds placement instructions in constraints between visually selected 3D surfaces. A solver gener￾ates geometrically feasible placements, followed by VLM-based se￾lection that assesses whether the resulting arrangements are plau￾sible in context. and geometric measurements help identify errors in size or place￾ment [LSD*26; HLC*26], turning a general visual mismatch into a concrete target for program revision. For editing, FirePlace [HBT*25] translates placement instruc￾tions into geometric constraints between object surfaces. It uses color-coded renderings to…
Figure 19 of the paper
Figure 19. DiagGen [CQX*26] uses simulated interaction to di￾agnose upstream construction errors. After generating deformable geometry and material properties from an image, it probes the asset through grasp-and-release episodes, routes repair to the responsi￾ble generation stage, and reruns that stage and its downstream de￾pendencies before retesting. and ParticleGen [ZYW*26] both translate visual diagnoses into pa￾rameter updates, but constrain those updates differently. ArtiMo plans causally ordered joint trajectories under URDF constraints and augments keyframes with motion cues so that a…

Showing 13 of the paper's 19 figures. Figures are the authors', reproduced from the paper.

References (291)
  1. [1] Huang, Zhening and Li, Yueyan and Chiu, Johnathan and Lyu, Xiaoyang and Zhou, Matt and Yao, Yuxin and Lasenby, Joan and Wu, Shangzhe , title =. 2026 , eprint = on Pith
  2. [2] Shi, Jin-Chuan and Ye, Binhong and Liu, Tao and He, Junzhe and Xu, Yangjinhui and Liu, Xiaoyang and Li, Zeju and Chen, Hao and Shen, Chunhua , title =. ACM SIGGRAPH 2026 Conference Papers , year =. doi:10.1145/3799902.3811134 , eprint = on Pithdoi:10.1145/3799902.3811134
  3. [3] Belouadi, Jonas and Boubekeur, Tamy and Kaiser, Adrien , title =. International Conference on Learning Representations , year =. 2509.22151 , eprinttype = on PithPith's reviewarXiv
  4. [4] Chen, Junhao and Li, Xiang and Ye, Xiaojun and Li, Chao and Fan, Zhaoxin and Zhao, Hao , title =. Proceedings of the 31st International Conference on Computational Linguistics , year =. 2404.04363 , eprinttype = on PithPith's reviewarXiv
  5. [5] Gao, Jin and Juluri, Saichandu , title =. 2026 , eprint = on Pith
  6. [6] Lam, Anson Y. and Li, Shuqing and Lyu, Michael R. , title =. 2026 , eprint = on Pith
  7. [7] Lin, Youtian and Yang, Yikang and Hu, Zhanpeng and Zhou, Mengqi and Zhang, Feihu and Cao, Xun and Liu, Jiaheng and Yao, Yao , title =. 2026 , eprint = on Pith
  8. [8] Pang, Bo and Pan, Jiaqi and Zhang, Xiaocheng and Xu, Jiacheng and Wang, Guoping and Wang, Peng-Shuai , title =. 2026 , eprint = on Pith
  9. [9] Zhang, Shuyuan and Jiang, Chenhan and Li, Zuoou and Deng, Jiankang , title =. Advances in Neural Information Processing Systems , year =. 2510.17603 , eprinttype = on PitharXiv
  10. [10] Ian Huang and Guandao Yang and Leonidas J. Guibas , title =. Computer Vision -- ECCV 2024 , pages =. 2024 , doi = on Pith
  11. [11] Yunqi Gu and Ian Huang and Jihyeon Je and Guandao Yang and Leonidas J. Guibas , title =. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , pages =. 2025 , doi = on Pith
  12. [12] Shaofeng Yin and Jiaxin Ge and Zora Zhiruo Wang and Chenyang Wang and Xiuyu Li and Michael J. Black and Trevor Darrell and Angjoo Kanazawa and Haiwen Feng , title =. arXiv preprint arXiv:2601.11109 , year = on PithPith's reviewarXiv
  13. [13] Xiaoxuan Ma and Jiashun Wang and Nicolas Ugrinovic and Yehonathan Litman and Kris Kitani , title =. arXiv preprint arXiv:2605.30338 , year = on PithPith's reviewarXiv
  14. [14] Zhengfei Kuang and Rui Lin and Long Zhao and Gordon Wetzstein and Saining Xie and Sanghyun Woo , title =. arXiv preprint arXiv:2512.22351 , year = on PitharXiv
  15. [15] Yipeng Gao and Lei Shu and Genzhi Ye and Xi Xiong and Ameesh Makadia and Meiqi Guo and Laurent Itti and Jindong Chen , title =. arXiv preprint arXiv:2606.01057 , year = on PithPith's reviewarXiv
  16. [16] Yuhao Du and Shunian Chen and Wenbo Zan and Peizhao Li and Mingxuan Wang and Dingjie Song and Bo Li and Yan Hu and Benyou Wang , title =. arXiv preprint arXiv:2412.14203 , year = on PithPith's reviewarXiv
  17. [17] Zhang, Yunzhi and Li, Zizhang and Zhou, Matt and Wu, Shangzhe and Wu, Jiajun , title =. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , year = on Pith
  18. [18] Matt Zhou and Ruining Li and Xiaoyang Lyu and Zhaomou Song and Zhening Huang and Chuanxia Zheng and Christian Rupprecht and Andrea Vedaldi and Shangzhe Wu , title =. arXiv preprint arXiv:2605.15187 , year = on PithPith's reviewarXiv
  19. [19] Mohammadmehdi Ataei and Farzaneh Askari and Kamal Rahimi Malekshan and Pradeep Kumar Jayaraman , title =. arXiv preprint arXiv:2604.24479 , year = on PithPith's reviewarXiv
  20. [20] Tian Xia and Tianrun Gao and Wenhao Deng and Long Wei and Xiaowei Qian and Yixian Jiang and Chenglei Yu and Tailin Wu , title =. arXiv preprint arXiv:2510.16559 , year = on PithPith's reviewarXiv
show 80 more
  1. [21] Hongchi Xia and Xuan Li and Zhaoshuo Li and Qianli Ma and Jiashu Xu and Ming-Yu Liu and Yin Cui and Tsung-Yi Lin and Wei-Chiu Ma and Shenlong Wang and Shuran Song and Fangyin Wei , title =. arXiv preprint arXiv:2602.10116 , year = on PitharXiv
  2. [22] Yi Zhang and Yunshuang Wang and Zeyu Zhang and Hao Tang , title =. arXiv preprint arXiv:2602.11757 , year = on PitharXiv
  3. [23] Parker Liu and Chenxin Li and Zhengxin Li and Yipeng Wu and Wuyang Li and Zhiqin Yang and Zhenyuan Zhang and Yunlong Lin and Sirui Han and Brandon Y. Feng , title =. arXiv preprint arXiv:2506.23329 , year = on PithPith's reviewarXiv
  4. [24] Lu Ling and Chen-Hsuan Lin and Tsung-Yi Lin and Yifan Ding and Yu Zeng and Yichen Sheng and Yunhao Ge and Ming-Yu Liu and Aniket Bera and Zhaoshuo Li , title =. International Conference on Learning Representations (ICLR) , year = on Pith
  5. [25] Jiacheng Chen and Ramin Mehran and Xuhui Jia and Saining Xie and Sanghyun Woo , title =. arXiv preprint arXiv:2506.17450 , year = on PithPith's reviewarXiv
  6. [26] Yandan Yang and Baoxiong Jia and Shujie Zhang and Siyuan Huang , title =. Advances in Neural Information Processing Systems (NeurIPS) , year = on Pith
  7. [27] Ian Huang and Yanan Bao and Karen Truong and Howard Zhou and Cordelia Schmid and Leonidas J. Guibas and Alireza Fathi , title =. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , pages =. 2025 , doi = on Pith
  8. [28] Sining Lu and Guan Chen and Nam Anh Dinh and Itai Lang and Ari Holtzman and Rana Hanocka , title =. arXiv preprint arXiv:2508.08228 , year = on PithPith's reviewarXiv
  9. [29] Ziniu Hu and Ahmet Iscen and Aashi Jain and Thomas Kipf and Yisong Yue and David A. Ross and Cordelia Schmid and Alireza Fathi , title =. International Conference on Machine Learning (ICML) , pages = on Pith
  10. [30] Giorgio Giannone and Anna Clare Doris and Amin Heyrani Nobari and Kai Xu and Akash Srivastava and Faez Ahmed , title =. arXiv preprint arXiv:2603.27448 , year = on PitharXiv
  11. [31] Jason Wei and Xuezhi Wang and Dale Schuurmans and Maarten Bosma and Brian Ichter and Fei Xia and Ed H. Chi and Quoc V. Le and Denny Zhou , title =. Advances in Neural Information Processing Systems (NeurIPS) , year = on Pith
  12. [32] Shunyu Yao and Jeffrey Zhao and Dian Yu and Nan Du and Izhak Shafran and Karthik Narasimhan and Yuan Cao , title =. International Conference on Learning Representations (ICLR) , year = on Pith
  13. [33] Shunyu Yao and Dian Yu and Jeffrey Zhao and Izhak Shafran and Thomas L. Griffiths and Yuan Cao and Karthik Narasimhan , title =. Advances in Neural Information Processing Systems (NeurIPS) , year = on Pith
  14. [34] Noah Shinn and Federico Cassano and Edward Berman and Ashwin Gopinath and Karthik Narasimhan and Shunyu Yao , title =. Advances in Neural Information Processing Systems (NeurIPS) , year = on Pith
  15. [35] Timo Schick and Jane Dwivedi-Yu and Roberto Dess. Toolformer: Language Models Can Teach Themselves to Use Tools , journal =. 2023 , eprint = on Pith
  16. [36] Shishir G. Patil and Tianjun Zhang and Xin Wang and Joseph E. Gonzalez , title =. Advances in Neural Information Processing Systems (NeurIPS) , year = on Pith
  17. [37] Yongliang Shen and Kaitao Song and Xu Tan and Dongsheng Li and Weiming Lu and Yueting Zhuang , title =. Advances in Neural Information Processing Systems (NeurIPS) , year = on Pith
  18. [38] Guanzhi Wang and Yuqi Xie and Yunfan Jiang and Ajay Mandlekar and Chaowei Xiao and Yuke Zhu and Linxi Fan and Anima Anandkumar , title =. Transactions on Machine Learning Research (TMLR) , year = on Pith
  19. [39] Pascal M. Procedural Modeling of Buildings , journal = on Pith
  20. [40] Weixi Feng and Wanrong Zhu and Tsu-jui Fu and Varun Jampani and Arjun Akula and Xuehai He and Sugato Basu and Xin Eric Wang and William Yang Wang , title =. Advances in Neural Information Processing Systems , volume =. 2023 , eprint = on Pith
  21. [41] Zhengyi Wang and Jonathan Lorraine and Yikai Wang and Hang Su and Jun Zhu and Sanja Fidler and Xiaohui Zeng , title =. arXiv preprint arXiv:2411.09595 , year = on PithPith's reviewarXiv
  22. [42] Juan A. Rodriguez and Abhay Puri and Shubham Agarwal and Issam H. Laradji and Pau Rodriguez and Sai Rajeswar and David Vazquez and Christopher Pal and Marco Pedersoli , title =. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , pages =.... on Pith
  23. [43] Jingyuan Chen and Sheng Jin and Haopeng Sun and Wentao Liu and Chen Qian , title =. arXiv preprint arXiv:2606.05058 , year = on PithPith's reviewarXiv
  24. [44] Yonglong Tian and Andrew Luo and Xingyuan Sun and Kevin Ellis and William T. Freeman and Joshua B. Tenenbaum and Jiajun Wu , title =. International Conference on Learning Representations (ICLR) , year = on Pith
  25. [45] R. Kenny Jones and Theresa Barton and Xianghao Xu and Kai Wang and Ellen Jiang and Paul Guerrero and Niloy J. Mitra and Daniel Ritchie , title =. ACM Transactions on Graphics (SIGGRAPH Asia) , volume =. 2020 , eprint = on Pith
  26. [46] Daniel Ritchie and Paul Guerrero and R. Kenny Jones and Niloy J. Mitra and Adriana Schulz and Karl D. D. Willis and Jiajun Wu , title =. Computer Graphics Forum (Eurographics STAR) , volume =. 2023 , eprint = on Pith
  27. [47] Carlos E. Jim. International Conference on Learning Representations (ICLR) , year = on Pith
  28. [48] John Yang and Carlos E. Jim. Advances in Neural Information Processing Systems (NeurIPS) , year = on Pith
  29. [49] Xingyao Wang and Boxuan Li and Yufan Song and Frank F. Xu and Xiangru Tang and Mingchen Zhuge and Jiayi Pan and Yueqi Song and Bowen Li and Jaskirat Singh and Hoang H. Tran and Fuqiang Li and Ren Ma and Mingzhang Zheng and Bill Qian and Yanjun Shao and Niklas Muennighoff and Y... on Pith
  30. [50] Wenting Zhao and Nan Jiang and Celine Lee and Justin T. Chiu and Claire Cardie and Matthias Gall. International Conference on Learning Representations (ICLR) , year =. 2412.01769 , eprinttype = on PithPith's reviewarXiv
  31. [51] Tianbao Xie and Danyang Zhang and Jixuan Chen and Xiaochuan Li and Siheng Zhao and Ruisheng Cao and Toh Jing Hua and Zhoujun Cheng and Dongchan Shin and Fangyu Lei and Yitao Liu and Yiheng Xu and Shuyan Zhou and Silvio Savarese and Caiming Xiong and Victor Zhong and Tao Yu , t... on Pith
  32. [52] Shuyan Zhou and Frank F. Xu and Hao Zhu and Xuhui Zhou and Robert Lo and Abishek Sridhar and Xianyi Cheng and Tianyue Ou and Yonatan Bisk and Daniel Fried and Uri Alon and Graham Neubig , title =. International Conference on Learning Representations (ICLR) , year = on Pith
  33. [53] Yiheng Xie and Towaki Takikawa and Shunsuke Saito and Or Litany and Shiqin Yan and Numair Khan and Federico Tombari and James Tompkin and Vincent Sitzmann and Srinath Sridhar , title =. Computer Graphics Forum , volume =. 2022 , pages = on Pith
  34. [54] Siddhartha Chaudhuri and Daniel Ritchie and Jiajun Wu and Kai Xu and Hao Zhang , title =. Computer Graphics Forum (Eurographics STAR) , volume =. 2020 , doi = on Pith
  35. [55] Xingyu Fu and Yushi Hu and Bangzheng Li and Yu Feng and Haoyu Wang and Xudong Lin and Dan Roth and Noah A. Smith and Wei-Chiu Ma and Ranjay Krishna , title =. European Conference on Computer Vision (ECCV) , year = on Pith
  36. [56] Shengbang Tong and Ellis Brown and Penghao Wu and Sanghyun Woo and Manoj Middepogu and Sai Charitha Akula and Jihan Yang and Shusheng Yang and Adithya Iyer and Xichen Pan and Ziteng Wang and Rob Fergus and Yann LeCun and Saining Xie , title =. Advances in Neural Information Pr... on Pith
  37. [57] Zhiheng Xi and Wenxiang Chen and Xin Guo and Wei He and Yiwen Ding and Boyang Hong and Ming Zhang and Junzhe Wang and Senjie Jin and Enyu Zhou and Rui Zheng and Xiaoran Fan and Xiao Wang and Limao Xiong and Yuhao Zhou and Weiran Wang and Changhao Jiang and Yicheng Zou and Xian... on PithPith's reviewarXiv
  38. [58] Junwei Liu and Kaixin Wang and Yixuan Chen and Xin Peng and Zhenpeng Chen and Lingming Zhang and Yiling Lou , title =. arXiv preprint arXiv:2409.02977 , year = on PithPith's reviewarXiv
  39. [59] Shuai Wang and Weiwen Liu and Jingxuan Chen and Yuqi Zhou and Weinan Gan and Xingshan Zeng and Yuhan Che and Shuai Yu and Xinlong Hao and Kun Shao and Bin Wang and Chuhan Wu and Yasheng Wang and Ruiming Tang and Jianye Hao , title =. arXiv preprint arXiv:2411.04890 , year = on PithPith's reviewarXiv
  40. [60] Chaoyun Zhang and Shilin He and Jiaxu Qian and Bowen Li and Liqun Li and Si Qin and Yu Kang and Minghua Ma and Guyue Liu and Qingwei Lin and Saravan Rajmohan and Dongmei Zhang and Qi Zhang , title =. arXiv preprint arXiv:2411.18279 , year = on PithPith's reviewarXiv
  41. [61] Dang Nguyen and Jian Chen and Yu Wang and Gang Wu and Namyong Park and Zhengmian Hu and Hanjia Lyu and Junda Wu and Ryan Aponte and Yu Xia and Xintong Li and Jing Shi and Hongjie Chen and Viet Dac Lai and Zhouhang Xie and Sungchul Kim and Ruiyi Zhang and Tong Yu and Mehrab Tan... on Pith
  42. [62] Robert L. Cook , title =. Proceedings of the 11th Annual Conference on Computer Graphics and Interactive Techniques (SIGGRAPH '84) , year = on Pith
  43. [63] Tzu-Mao Li and Miika Aittala and Fr. Differentiable. ACM Transactions on Graphics (SIGGRAPH Asia) , volume = on Pith
  44. [64] Samuli Laine and Janne Hellsten and Tero Karras and Yeongho Seol and Jaakko Lehtinen and Timo Aila , title =. ACM Transactions on Graphics (SIGGRAPH Asia) , volume =. 2020 , eprint = on Pith
  45. [65] David Baraff and Andrew Witkin , title =. Proceedings of the 25th Annual Conference on Computer Graphics and Interactive Techniques (SIGGRAPH '98) , year = on Pith
  46. [66] Jos Stam , title =. Proceedings of the 26th Annual Conference on Computer Graphics and Interactive Techniques (SIGGRAPH '99) , year = on Pith
  47. [67] Yuanming Hu and Tzu-Mao Li and Luke Anderson and Jonathan Ragan-Kelley and Fr. ACM Transactions on Graphics (SIGGRAPH Asia) , volume =. 2019 , doi = on Pith
  48. [68] arXiv preprint arXiv:2303.08774 , year = on PithPith's reviewarXiv
  49. [69] Haotian Liu and Chunyuan Li and Qingyang Wu and Yong Jae Lee , title =. Advances in Neural Information Processing Systems (NeurIPS) , year = on Pith
  50. [70] Mark Chen and Jerry Tworek and Heewoo Jun and Qiming Yuan and Henrique Ponde de Oliveira Pinto and Jared Kaplan and Harri Edwards and Yuri Burda and Nicholas Joseph and Greg Brockman and Alex Ray and Raul Puri and Gretchen Krueger and Michael Petrov and Heidy Khlaaf and Girish... on PithPith's reviewarXiv
  51. [71] arXiv preprint arXiv:2403.05530 , year = on PithPith's reviewarXiv
  52. [72] arXiv preprint arXiv:2412.15115 , year =. 2412.15115 , eprinttype = on PithPith's reviewarXiv
  53. [73] Peng Wang and Shuai Bai and Sinan Tan and Shijie Wang and Zhihao Fan and Jinze Bai and Keqin Chen and Xuejing Liu and Jialin Wang and Wenbin Ge and Yang Fan and Kai Dang and Mengfei Du and Xuancheng Ren and Rui Men and Dayiheng Liu and Chang Zhou and Jingren Zhou and Junyang L... on PithPith's reviewarXiv
  54. [74] arXiv preprint arXiv:2412.19437 , year = on PithPith's reviewarXiv
  55. [75] arXiv preprint arXiv:2501.12599 , year = on PithPith's reviewarXiv
  56. [76] arXiv preprint arXiv:2504.07491 , year = on PithPith's reviewarXiv
  57. [77] CADIR: A Cross-Backend Editable Intermediate Representation for Agentic CAD Generation , author=. arXiv preprint arXiv:2608.00891 , year= on PithPith's reviewarXiv
  58. [78] ParticleGen: A Multi-Agent System for Particle Effects Generation , author=. arXiv preprint arXiv:2608.00629 , year= on PithPith's reviewarXiv
  59. [79] GS-Agent: Creating 4D Physical Worlds With Generative Simulation , author=. arXiv preprint arXiv:2607.21522 , year= on PithPith's reviewarXiv
  60. [80] MUSE: Agentic 3D Scene Authoring via Memory-Grounded Incremental Requirement Satisfaction , author=. arXiv preprint arXiv:2606.14168 , year= on PitharXiv
  61. [81] SimWorlds: A Multi-Agent System for Dynamic 3D Scene Creation , author=. arXiv preprint arXiv:2607.01766 , year= on PithPith's reviewarXiv
  62. [82] SceneConductor: 3D Scene Generation from a Single Image with Multi-Agent Orchestration , author=. arXiv preprint arXiv:2606.08402 , year= on PithPith's reviewarXiv
  63. [83] SimFoundry: Modular and Automated Scene Generation for Policy Learning and Evaluation , author=. arXiv preprint arXiv:2606.28276 , year= on PithPith's reviewarXiv
  64. [84] Yuyang Yin and Zixiang Li and Longxuan Deng and Hongkai Li and Shifang Zhao and Junnan Liu and Weirong Huang and Mengyu Wang and Tianxiao Fu and Yikai Wang and Peng-Shuai Wang and Xiaojie Jin and Yao Zhao and Yunchao Wei , title =. arXiv preprint arXiv:2608.12314 , year = on PithPith's reviewarXiv
  65. [85] Tsz Hei Fan and Choi Wing Fung and Yuxuan Wan and Shuqing Li and Michael R. Lyu , title =. arXiv preprint arXiv:2607.11594 , year = on PithPith's reviewarXiv
  66. [86] Nicholas Pfaff and Thomas Cohn and Sergey Zakharov and Rick Cory and Russ Tedrake , title =. International Conference on Machine Learning (ICML) , year = on Pith
  67. [87] Yian Wang and Han Yang and Minghao Guo and Xiaowen Qiu and Tsun-Hsuan Wang and Wojciech Matusik and Joshua B. Tenenbaum and Chuang Gan , title =. International Conference on Learning Representations (ICLR) , year = on Pith
  68. [88] Lei Yin and Wentao Cheng and Zhida Qin and Tianyu Huang and Yidong Li and Gangyi Ding , title =. Findings of the Association for Computational Linguistics: ACL 2026 , pages =. 2026 , publisher = on Pith
  69. [89] Ziya Erko. arXiv preprint arXiv:2603.19708 , year = on PitharXiv
  70. [91] Global-local tree search in vlms for 3d indoor scene generation , author=. 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , pages=. 2025 , organization= on Pith
  71. [92] Yankuan Chi and Xiang Li and Zixuan Huang and James Matthew Rehg , title =. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , pages = on Pith
  72. [93] Saugat Adhikari and Ashok Prasad Neupane and Pramish Paudel and Ajad Chhatkuli and Danda Pani Paudel , title =. arXiv preprint arXiv:2608.06161 , year = on PithPith's reviewarXiv
  73. [94] Rui-Huan Wang and Si-Tong Wei and Jia-Qi He and Heng-Yi Wei and Baoquan Chen and Peng-Shuai Wang , title =. arXiv preprint arXiv:2608.17975 , year = on PithPith's reviewarXiv
  74. [95] Yixuan Yang and Zhen Luo and Wanshui Gan and Jinkun Hao and Junru Lu and Jinghao Yan and Zhaoyang Lyu and Xudong Xu , title =. arXiv preprint arXiv:2605.18451 , year = on PithPith's reviewarXiv
  75. [96] Hao Wang and Wenhui Zhu and Shao Tang and Zhipeng Wang and Xuanzhao Dong and Xin Li and Xiwen Chen and Ashish Bastola and Xinhao Huang and Yalin Wang and Abolfazl Razi , title =. arXiv preprint arXiv:2601.07143 , year = on PitharXiv
  76. [97] Yutaro Yamada and Khyathi Chandu and Bill Yuchen Lin and Jack Hessel and Ilker Yildirim and Yejin Choi , title =. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (System Dem... on PithPith's reviewarXiv
  77. [98] 3d-gpt: Procedural 3d modeling with large language models , author=. 2025 International Conference on 3D Vision (3DV) , pages=. 2025 , organization= on Pith
  78. [99] Scenegenagent: Precise industrial scene generation with coding agent , author=. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) , pages= on Pith
  79. [100] Dimitrios Mallis and Ahmet Serdar Karadeniz and Sebastian Cavada and Danila Rukhovich and Niki Foteinopoulou and Kseniya Cherenkova and Anis Kacem and Djamila Aouada , title =. Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) , year =. 2412.13810 ... on PithPith's reviewarXiv
  80. [101] Elias Berger and Muhammad Usama and Jan Mehlst. Physics-in-the-Loop: A Hybrid Agentic Architecture for Validated. arXiv preprint arXiv:2605.19717 , year = on PithPith's reviewarXiv

The first 100 of 291 references are listed here. All of them are in this paper's citation record.

Challenges and discussion

Disagree with the review, or know something it missed? Leave a comment, or challenge this review formally: a challenge becomes a permanent entry in this paper's record, and its decision is published here.

No comments yet.

Sign in to comment