Pith. sign in

Paper Citation Record · LEDGER

CARV: A Diagnostic Benchmark for Compositional Analogical Reasoning in Multimodal LLMs

As of 8 August 2026, this Paper Citation Record lists 12 of 12 outbound references and 0 inbound Pith citation observations for arXiv:2603.27958.

A citation records a reference. It does not transfer a finding from one paper to another.

pith.paper-citation-record.v1
2603.27958 v2

Coverage vector

measured 12 of 12 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-02T17:14:02.051087Z

measured 12 of 12 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-07T06:34:17.273281+00:00

measured 0 of 0 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links

measured 0 of 1 external citation measurements

A source-named dated measurement, never combined with another source.

Source: cited_works

Reference resolution

12 of 12 outbound references displayed

  • verified exact0
  • verified fuzzy0
  • unresolved12
  • parse uncertain0
  • malformed identifier0
  • metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 0148a464-eb11-419d-adf7-0db6dcd5eb62 · outbound

This paper cites an unresolved cited work.

CARV: A Diagnostic Benchmark for Compositional Analogical Reasoning in Multimodal LLMs Unresolved cited work

Reference 1

Resolution
unresolved
no resolver link, observed 2026-08-02T17:14:02.007940Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-02T17:14:02.007940Z digest=sha256:66dea464bd796a92eab6c899c78d1e423305d087ed17109072fd07b72982c010

Observation e6d98273-3616-4f54-ae4e-2f47de107e88 · outbound

This paper cites [Quantity] changes from two knives to one knife.

CARV: A Diagnostic Benchmark for Compositional Analogical Reasoning in Multimodal LLMs [Quantity] changes from two knives to one knife

Reference 2

Resolution
unresolved
no resolver link, observed 2026-08-02T17:14:01.993713Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-02T17:14:01.993713Z digest=sha256:ee2e9106492d507a2dcf59fbc9469fab92df0b06be10e58957a7e3b6ffbfbf09

Observation baa2aba9-4873-417a-87aa-edf5060c0d4c · outbound

This paper cites [Color] changes from wooden to red.[Quantity] should be at most two knives, as a change from two to one is also possible.

CARV: A Diagnostic Benchmark for Compositional Analogical Reasoning in Multimodal LLMs [Color] changes from wooden to red.[Quantity] should be at most two knives, as a change from two to one is also possible

Reference 3

Resolution
unresolved
no resolver link, observed 2026-08-02T17:14:02.000742Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-02T17:14:02.000742Z digest=sha256:d9af9110b587946d040d92463c6b11c51582b77bf55d5d2a92e18d4a6f573d6a

Observation 04690425-6ec8-48ff-9664-de267ffbe089 · outbound

This paper cites The final output should be a concise caption of the resulting image in JSON format.

CARV: A Diagnostic Benchmark for Compositional Analogical Reasoning in Multimodal LLMs The final output should be a concise caption of the resulting image in JSON format

Reference 4

Resolution
unresolved
no resolver link, observed 2026-08-02T17:14:02.025816Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-02T17:14:02.025816Z digest=sha256:405051faaf1b619b218759bbb06116a8c359015afede79e173e4179e95001aca

Observation 207b1f9d-17c6-4b0e-925e-8aadcc23a89a · outbound

This paper cites Describe what changes, and from which to which.

CARV: A Diagnostic Benchmark for Compositional Analogical Reasoning in Multimodal LLMs Describe what changes, and from which to which

Reference 6

Resolution
unresolved
no resolver link, observed 2026-08-02T17:14:02.013248Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-02T17:14:02.013248Z digest=sha256:4bf5b11dee60f47c8da63499bba37f7c07f6502b25e0fa50a0bd20960b83dd28

Observation 0cc2df8f-58c0-4b72-8f8a-8a618eb83e4a · outbound

This paper cites Qwen3 Technical Report.

CARV: A Diagnostic Benchmark for Compositional Analogical Reasoning in Multimodal LLMs Qwen3 Technical Report

Reference 7

Resolution
unresolved
no resolver link, observed 2026-08-02T17:14:01.969941Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-02T17:14:01.969941Z digest=sha256:7cc912ac6e6b667cd6ec30cfdd593d4d3d1555b8f42f496de1facb2c795455b2

Observation 10a4dd22-9d3c-4a61-9d41-458678ab9add · outbound

This paper cites description When T1 or T2 contains items which indicate no change For example, T1 includes ”[property] remains the same”, and T2 includes ”[property] changes from A to B”.

CARV: A Diagnostic Benchmark for Compositional Analogical Reasoning in Multimodal LLMs description When T1 or T2 contains items which indicate no change For example, T1 includes ”[property] remains the same”, and T2 includes ”[property] changes from A to B”

Reference 8

Resolution
unresolved
no resolver link, observed 2026-08-02T17:14:02.019471Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-02T17:14:02.019471Z digest=sha256:916b36297014db6944015af506953c543c014e4a4487386cc7803ad8efb93014

Observation 1b268db8-71cb-49a7-b003-7af9f2975dd2 · outbound

This paper cites Instruction: (1) whether the response is correct according to the reference captions.

CARV: A Diagnostic Benchmark for Compositional Analogical Reasoning in Multimodal LLMs Instruction: (1) whether the response is correct according to the reference captions

Reference 9

Resolution
unresolved
no resolver link, observed 2026-08-02T17:14:02.031282Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-02T17:14:02.031282Z digest=sha256:a434f7ab3822a17d7bfec69e0bf3205ad156c67f10c85ff1e06dd977984fb1bf

Observation ed645a68-296a-4427-b455-0a57de74966d · outbound

This paper cites Instruction: (1) When describing one transformation, the model must follow [property] changes from [value A] to [value B].

CARV: A Diagnostic Benchmark for Compositional Analogical Reasoning in Multimodal LLMs Instruction: (1) When describing one transformation, the model must follow [property] changes from [value A] to [value B]

Reference 10

Resolution
unresolved
no resolver link, observed 2026-08-02T17:14:02.037343Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-02T17:14:02.037343Z digest=sha256:1eaf435b5bd985dd738b99d39acd098a600127012e5fcf968c43aa7fe4a0ec6f

Observation e47d001b-8402-4c15-8598-668888f5874c · outbound

This paper cites Reference target transformation: Transformation T: {target transformation}.

CARV: A Diagnostic Benchmark for Compositional Analogical Reasoning in Multimodal LLMs Reference target transformation: Transformation T: {target transformation}

Reference 11

Resolution
unresolved
no resolver link, observed 2026-08-02T17:14:02.044849Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-02T17:14:02.044849Z digest=sha256:cbb3c07f62b4f87903a3ad28da7b50e932fd36a54d4139b53390a0e29946074d

Observation 1c5da054-e35e-4312-9c40-19f020a4884b · outbound

This paper cites an unresolved cited work.

CARV: A Diagnostic Benchmark for Compositional Analogical Reasoning in Multimodal LLMs Unresolved cited work

Reference 12

Resolution
unresolved
no resolver link, observed 2026-08-02T17:14:02.051087Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-02T17:14:02.051087Z digest=sha256:a84320a916b7e7688a16c8b27ecb2a0e1720f1f694b075971f942af6f651fc77

Observation 4a7ecbc1-5942-4824-b80a-9e3f101176a4 · outbound

This paper cites Compositional Scene Understanding through Inverse Generative Modeling.

CARV: A Diagnostic Benchmark for Compositional Analogical Reasoning in Multimodal LLMs Compositional Scene Understanding through Inverse Generative Modeling

Reference 280

Resolution
unresolved
no resolver link, observed 2026-08-02T17:14:01.979367Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-02T17:14:01.979367Z digest=sha256:a22954cd84b522bbe5f01e212f2cb20c013759901c1446a654aed7f8118d072a

Pith citing papers

No inbound Pith citation observations are available.