Pith. sign in

Paper Citation Record · LEDGER

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit

As of 10 August 2026, this Paper Citation Record lists 57 of 57 outbound references and 4 inbound Pith citation observations for arXiv:2501.09240.

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

pith.paper-citation-record.v1
2501.09240 v1

Coverage vector

measured 57 of 57 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-10T20:15:28.491785Z

measured 61 of 61 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-10T06:31:04.303077+00:00

measured 4 of 4 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-07T11:12:59.912293Z

measured 0 of 1 external citation measurements

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

Source: arxiv_reference, observed 2026-05-11T16:46:15.163748Z

Reference resolution

57 of 57 outbound references displayed

  • verified exact2
  • verified fuzzy9
  • unresolved46
  • parse uncertain0
  • malformed identifier0
  • metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 751bc24e-2aa5-47c3-91fa-b57bb0a59825 · outbound

This paper cites What learning algorithm is in-context learning? Investigations with linear models.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit What learning algorithm is in-context learning? Investigations with linear models

Reference 1

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.218515Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.218515Z digest=sha256:80a49abbd849332f493ca7a4551c06c778111c2734d8a5bc55f0a5434eca76ab

Observation 850b2601-5e83-46c8-bc7d-4735c84a836f · outbound

This paper cites maps-to” token (->). Conversely, in the trained-from-scratch transformer, task information is stored in the y token (i.e., the “label.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit maps-to” token (->). Conversely, in the trained-from-scratch transformer, task information is stored in the y token (i.e., the “label

Reference 3

Resolution
verified fuzzy
raw_fallback, observed 2026-08-10T20:15:29.254604Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

source=pdf_text observed=2026-08-10T20:15:28.477168Z digest=sha256:239d49a6e06a1322f25d352be7c711794f54e2fe6ab8ee9dba912d1fd7eeb605

Observation 506fe749-b885-4c55-9b4a-9d50976e33e2 · outbound

This paper cites Task Prompt Vectors: Effective Initialization through Multi-Task Soft-Prompt Transfer.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Task Prompt Vectors: Effective Initialization through Multi-Task Soft-Prompt Transfer

Reference 4

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.233830Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.233830Z digest=sha256:67f8add7676537956a2f6739264e2457c9058ccd0ca075eca4a9231b1f40aae3

Observation 0f972f5c-7f52-4b72-abae-c4f76631f33f · outbound

This paper cites Longformer: The Long-Document Transformer.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Longformer: The Long-Document Transformer

Reference 5

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.238548Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.238548Z digest=sha256:2cb2f15e81458f957fca3fe82315970046f8c931ab3302ec45a328230da6fd46

Observation 841a9300-336d-4acb-ae70-666176c2fd9d · outbound

This paper cites Transformers generalize differently from information stored in context vs in weights.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Transformers generalize differently from information stored in context vs in weights

Reference 8

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.253212Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.253212Z digest=sha256:70e7b51cae27ff88e81dfa1bc2efbf6f21be631cf63ee16ae97a3d14c0c436b4

Observation c0d1b830-fcaa-4f2e-bac8-2c4f1a01e5b5 · outbound

This paper cites Analyzing Transformers in Embedding Space.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Analyzing Transformers in Embedding Space

Reference 9

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.258738Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.258738Z digest=sha256:845aacf4e79b899c397b38b0ae91f112fdc92acbba6353e03c92ac2029b96cb2

Observation 14115c38-ca1e-40de-b3ef-f4c7601b3faa · outbound

This paper cites What Can Transformers Learn In-Context? A Case Study of Simple Function Classes.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit What Can Transformers Learn In-Context? A Case Study of Simple Function Classes

Reference 11

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.269173Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.269173Z digest=sha256:be9aef7359ec974b649f0dcecb4445b2714c8f6a79bc71c295b7df2eeedf44ac

Observation 8fd0073d-c676-4078-9e1d-3d58c5ee5232 · outbound

This paper cites Transformer Feed-Forward Layers Build Predictions by Promoting Concepts in the Vocabulary Space.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Transformer Feed-Forward Layers Build Predictions by Promoting Concepts in the Vocabulary Space

Reference 12

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.274332Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.274332Z digest=sha256:681e5d61b2788441fe7d32df9fa7eaa9f5dd325bbacb559cd04db02f74fe0db4

Observation 93094e9b-4986-42cb-aa0d-c6069f91e9f1 · outbound

This paper cites Think before you speak: Training Language Models With Pause Tokens.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Think before you speak: Training Language Models With Pause Tokens

Reference 13

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.279307Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.279307Z digest=sha256:ed4b8015785433d8fafb3c73c05fd68f3a3f8130f91439bd61d246b66b018d0f

Observation 28b1809c-cb9a-4666-954e-0b782108759e · outbound

This paper cites Have Faith in Faithfulness: Going Beyond Circuit Overlap When Finding Model Mechanisms.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Have Faith in Faithfulness: Going Beyond Circuit Overlap When Finding Model Mechanisms

Reference 14

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.284054Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.284054Z digest=sha256:1d694b8d97bb46a2b63560e072b024664d610dba3b741c0ed9cc99d3001202a2

Observation 5e6f5a75-19d4-4de5-8136-858dbaadc4b7 · outbound

This paper cites In-Context Learning Creates Task Vectors.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit In-Context Learning Creates Task Vectors

Reference 15

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.289415Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.289415Z digest=sha256:09386730fa5b1bd254e96be286e6fcd7d87f5c4154a8affaa4b22e5023487dcf

Observation 03e4a715-42ac-4bd5-8044-7b4a208afca8 · outbound

This paper cites Finding Visual Task Vectors.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Finding Visual Task Vectors

Reference 16

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.294418Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.294418Z digest=sha256:6bfc065a77c1e94951ea80e742030749b5acc4ef5c1ef723102a52f29f237821

Observation 1c13ee11-352d-4704-8f07-83e07c774448 · outbound

This paper cites Editing Models with Task Arithmetic.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Editing Models with Task Arithmetic

Reference 17

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.299939Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.299939Z digest=sha256:d85360464ee41e6b2660bae45706fd79457fd0d61e8e9def1b17a3666da7f612

Observation dd88c99d-ac3b-45cb-b924-01d6d504a4d7 · outbound

This paper cites semanticscholar.org/CorpusID:254408495.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit semanticscholar.org/CorpusID:254408495

Reference 18

Resolution
verified fuzzy
raw_fallback, observed 2026-08-10T20:15:29.357909Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

source=pdf_text observed=2026-08-10T20:15:28.305590Z digest=sha256:fca439b5ff79afb97b34a554e65b0cf9653634114bcfeb4a502284da3ed4e164

Observation d08baaec-bca3-4d8d-aea3-13d44dd243c5 · outbound

This paper cites org/CorpusID:271741273.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit org/CorpusID:271741273

Reference 19

Resolution
verified fuzzy
raw_fallback, observed 2026-08-10T20:15:29.344147Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

source=pdf_text observed=2026-08-10T20:15:28.310118Z digest=sha256:ebed9f2e7059ead39133c649582961d4b4f982b5acf68ca23207f189044cb1a6

Observation d012e0be-3c05-401a-bcf4-145b16bd7d33 · outbound

This paper cites Ground-Truth Labels Matter: A Deeper Look into Input-Label Demonstrations.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Ground-Truth Labels Matter: A Deeper Look into Input-Label Demonstrations

Reference 20

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.314561Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.314561Z digest=sha256:93e6efa3dadd047de35aa8578499dcbf54e8e427daaa489e6031e5796deb93b3

Observation cba17401-f06e-4235-9bcc-1dc4a7547050 · outbound

This paper cites When can transformers compositionally generalize in-context?.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit When can transformers compositionally generalize in-context?

Reference 21

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.319086Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.319086Z digest=sha256:a4a233d3dfcec88ec31fc5e8d894b9917b4a62c159ec3f0699a4af2ab84c321a

Observation 30666b6b-eab3-4d8e-a4c7-c9fb3254bc55 · outbound

This paper cites Supervised Pretraining Can Learn In-Context Reinforcement Learning.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Supervised Pretraining Can Learn In-Context Reinforcement Learning

Reference 22

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.324021Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.324021Z digest=sha256:d156b102c64ee00c19dc4075cd0ed8a651a60f950aa07183d863f583a1a661d1

Observation b8b05d93-2c5f-4ad7-8448-42d58bac416e · outbound

This paper cites Implicit In-context Learning.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Implicit In-context Learning

Reference 23

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.328902Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.328902Z digest=sha256:97c85550acf605a4075969a053cbbdc32c9b18a9d150a3eeee91cf7ceb1e8c0e

Observation 08e540c3-fbe6-485c-acf1-2c5bbfe9de63 · outbound

This paper cites Dual Operating Modes of In-Context Learning.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Dual Operating Modes of In-Context Learning

Reference 24

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.333268Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.333268Z digest=sha256:8992e126161338aca07e9e39e65e62ad7196c0feded600b12e6a2e9ff9351271

Observation edfe781f-5283-43f7-8127-8e5d2e6c7c74 · outbound

This paper cites In-context Vectors: Making In Context Learning More Effective and Controllable Through Latent Space Steering.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit In-context Vectors: Making In Context Learning More Effective and Controllable Through Latent Space Steering

Reference 25

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.337375Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.337375Z digest=sha256:4290bd97d527e87bd7cf55266aa24d5f98ea59159772166203b7fd2c1e33879a

Observation e80988d6-b352-4ab8-b293-628c65981c70 · outbound

This paper cites GPT Understands, Too.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit GPT Understands, Too

Reference 26

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.341830Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.341830Z digest=sha256:cc9e406e9b81a9be9ece6abe1122f8f73c360800f4e480f17b01562a71e7cbaf

Observation fe283fdf-e15f-490c-b50c-dac1fb675ae5 · outbound

This paper cites Does In-Context Learning Really Learn? Rethinking How Large Language Models Respond and Solve Tasks via In-Context Learning.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Does In-Context Learning Really Learn? Rethinking How Large Language Models Respond and Solve Tasks via In-Context Learning

Reference 27

Resolution
verified exact
local_arxiv, observed 2026-08-10T20:15:28.841850Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

source=pdf_text observed=2026-08-10T20:15:28.346500Z digest=sha256:21b9acb137a4980b7939210b7a724d63a3cfe621c9cf1e8d7426f5b1c7e0459f

Observation 65e0bc34-bfc1-4038-ba68-69cda10bf560 · outbound

This paper cites Sparser is Faster and Less is More: Efficient Sparse Attention for Long-Range Transformers.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Sparser is Faster and Less is More: Efficient Sparse Attention for Long-Range Transformers

Reference 28

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.350805Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.350805Z digest=sha256:0c6bb1cbf3ebb4e2a56cd4621059e8116eda7f3208d0ef0b400817ec3eca77bd

Observation 4fdfe0d7-6034-46b5-8b17-7b01cfe871b7 · outbound

This paper cites Language Models Implement Simple Word2Vec-style Vector Arithmetic.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Language Models Implement Simple Word2Vec-style Vector Arithmetic

Reference 29

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.355177Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.355177Z digest=sha256:aa1a588a4e4bb03b2fdcdac7ff60ed71f65c19b610bfa22aa876e9f426b13213

Observation f72d7590-7a4c-4ee1-963e-25583ad2b92f · outbound

This paper cites Does learning the right latent variables necessarily improve in-context learning?.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Does learning the right latent variables necessarily improve in-context learning?

Reference 30

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.359605Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.359605Z digest=sha256:37f0c7a90946790ab6167419dd5657edbae979570a90a69d5332da3e504182f7

Observation 3c68c9e0-af4e-464d-8e8a-711f72b5215a · outbound

This paper cites Learning to Compress Prompts with Gist Tokens.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Learning to Compress Prompts with Gist Tokens

Reference 31

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.364124Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.364124Z digest=sha256:431ac48f0510beaf14d9a327902d9c7209243e60d0fad35baeb0d5b3d6957efa

Observation 68f38ecf-1787-4a1f-ab41-b07864d97b77 · outbound

This paper cites Transformers Can Do Bayesian Inference.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Transformers Can Do Bayesian Inference

Reference 32

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.368920Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.368920Z digest=sha256:febbc738ceb03501f989fd6de0a44eb34ed63f63725f87394bd020c916c9375d

Observation 0e782682-5566-44d5-ac41-16fd64e0b938 · outbound

This paper cites LIVE: Learnable In-Context Vector for Visual Question Answering.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit LIVE: Learnable In-Context Vector for Visual Question Answering

Reference 33

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.373592Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.373592Z digest=sha256:9fc1747ad018fc3b0aa4fbdeb9170016f70bc2440b60c526b24e41089e6010d4

Observation fedd93a1-5549-499d-a5e3-96a7bd6e5f2b · outbound

This paper cites Robust Concept Erasure Using Task Vectors.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Robust Concept Erasure Using Task Vectors

Reference 34

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.378401Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.378401Z digest=sha256:bf699b4db0470701479b00381ccbca6453d44b07c3914459cd2fa31d404604be

Observation 3946ad61-eec8-46ad-8922-e2373645740c · outbound

This paper cites Investigating the Effectiveness of HyperTuning via Gisting.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Investigating the Effectiveness of HyperTuning via Gisting

Reference 35

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.383039Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.383039Z digest=sha256:ce9d2e50f194c7ecc84f0423dd89fd5e887fb9afdba0933d30e89fef01b9be6f

Observation c2fe5417-dbc2-40c8-9213-8004bfb96902 · outbound

This paper cites Blockwise Self-Attention for Long Document Understanding.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Blockwise Self-Attention for Long Document Understanding

Reference 36

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.388054Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.388054Z digest=sha256:428928ebb90e4f8a74ef0ac3f8b31ec2ca72ded2d29cfeee197f0c80c4bcc6e6

Observation abeebd98-5226-414a-a373-963e6a819852 · outbound

This paper cites Discovering modular solutions that generalize compositionally.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Discovering modular solutions that generalize compositionally

Reference 38

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.398835Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.398835Z digest=sha256:56d6825d0573244543b05d45aa38e3f4c42b7e4315a3bbe5a6ff9dc2915d28a8

Observation 35c4bf51-4627-4fb5-a0dc-a221107076f9 · outbound

This paper cites Where does In-context Translation Happen in Large Language Models.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Where does In-context Translation Happen in Large Language Models

Reference 39

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.403519Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.403519Z digest=sha256:38e4114d3e20c643a8326d6189964d8d8c5b5d2c4ec20c352495d8a8557ab680

Observation eb54ad68-6d35-497b-adf4-12d5912b13ab · outbound

This paper cites What needs to go right for an induction head? A mechanistic study of in-context learning circuits and their formation.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit What needs to go right for an induction head? A mechanistic study of in-context learning circuits and their formation

Reference 40

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.407980Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.407980Z digest=sha256:7fc453c4f175e775f96f9302fbd57b27a1addc3e5ca90dafb5bb1137d6bc4c9c

Observation 7e428b76-947d-432e-a27e-0f1b496cbb12 · outbound

This paper cites Function Vectors in Large Language Models.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Function Vectors in Large Language Models

Reference 41

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.412257Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.412257Z digest=sha256:3d72c8b75e2155f119024a3cb5f7efbe18992dfb0ba222db4d96bdb334c5e60e

Observation 25c9cb9a-76ff-478a-af77-ef1deba04bb8 · outbound

This paper cites Transformers learn in-context by gradient descent.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Transformers learn in-context by gradient descent

Reference 42

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.417098Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.417098Z digest=sha256:3dde0dce28e8cc593505fa55fa554df52ff66c000c3898c4abf7d3409390944b

Observation 59fecf7f-02a7-4741-806d-58087e8e0ca2 · outbound

This paper cites Uncovering mesa-optimization algorithms in Transformers.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Uncovering mesa-optimization algorithms in Transformers

Reference 43

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.421835Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.421835Z digest=sha256:ea7ed8d9c0955753f65c6a789646a96e1c5e1b3953156c46925981a8da69975b

Observation fb34adb7-acd4-4b6a-83e0-145a026b5f36 · outbound

This paper cites Interpretability in the Wild: a Circuit for Indirect Object Identification in GPT-2 small.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Interpretability in the Wild: a Circuit for Indirect Object Identification in GPT-2 small

Reference 44

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.427015Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.427015Z digest=sha256:294d0117418e163874d21d1290af0767acd28fb464f095d839b06bf5e9716f87

Observation ec916408-9287-4198-8d65-9e4a542059f3 · outbound

This paper cites An Explanation of In-context Learning as Implicit Bayesian Inference.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit An Explanation of In-context Learning as Implicit Bayesian Inference

Reference 45

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.431801Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.431801Z digest=sha256:0433b9953f360b5e4e21efc49889140fb6ef5d69bcc892864e6ccb65ed5ce309

Observation f40bfb9a-f028-4551-afc0-58de1a7a40e9 · outbound

This paper cites Compress, Then Prompt: Improving Accuracy-Efficiency Trade-off of LLM Inference with Transferable Prompt.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Compress, Then Prompt: Improving Accuracy-Efficiency Trade-off of LLM Inference with Transferable Prompt

Reference 46

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.436609Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.436609Z digest=sha256:ee19323265f2494ef119109ec12739af86c45ee7fcb632951c190309dad71995

Observation 64e2e645-84ee-4e26-b5bd-0fc051f47ffd · outbound

This paper cites Pretraining Data Mixtures Enable Narrow Model Selection Capabilities in Transformer Models.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Pretraining Data Mixtures Enable Narrow Model Selection Capabilities in Transformer Models

Reference 47

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.441288Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.441288Z digest=sha256:1c4816cc245c539633dcac87f88fdc5922812dc0e7d0a9fd54b862662aedf50b

Observation 02445c5f-758c-4a37-85e8-84303ed66770 · outbound

This paper cites Scaling Up Personalized Image Aesthetic Assessment via Task Vector Customization.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Scaling Up Personalized Image Aesthetic Assessment via Task Vector Customization

Reference 48

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.446887Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.446887Z digest=sha256:f187a168c33769dfa577df4e91f0dbdfeaeb3c46f4cdb45fe581126268cd8c72

Observation 0003d98f-9f8d-4703-9096-be1992e0f96f · outbound

This paper cites What and How does In-Context Learning Learn? Bayesian Model Averaging, Parameterization, and Generalization.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit What and How does In-Context Learning Learn? Bayesian Model Averaging, Parameterization, and Generalization

Reference 49

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.451307Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.451307Z digest=sha256:a25cb73e557cba641ce9fd7339bc6bc05bd1f3e278bf97c7104249cd2fba5bb4

Observation a2f513ee-ae80-4cfd-b423-e1e739536e8e · outbound

This paper cites learning.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit learning

Reference 50

Resolution
verified fuzzy
raw_fallback, observed 2026-08-10T20:15:29.315708Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

source=pdf_text observed=2026-08-10T20:15:28.457443Z digest=sha256:7a81bf18f383546da6f34a5ac02b63d9dac319897f17fcb75f19d1e06efd3b84

Observation 5641d13d-503b-4d9b-a70e-4a4d6ff96e11 · outbound

This paper cites function vector,.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit function vector,

Reference 51

Resolution
verified fuzzy
raw_fallback, observed 2026-08-10T20:15:29.301122Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

source=pdf_text observed=2026-08-10T20:15:28.462769Z digest=sha256:ef3eda1254fab25ef701b65efd6811d64ac5457e7d477326abe682b15d6692ec

Observation 44b61a1d-988e-48c9-95cb-c358301e63a4 · outbound

This paper cites Beyond activations, task information can also be encoded in a task token Bai et al.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Beyond activations, task information can also be encoded in a task token Bai et al

Reference 52

Resolution
verified fuzzy
raw_fallback, observed 2026-08-10T20:15:29.284534Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

source=pdf_text observed=2026-08-10T20:15:28.468012Z digest=sha256:3bb246c6be37fd2d7abfa48b195d9334609b538991db0200d48e5655790e2d88

Observation e94eec66-f2de-4c2f-9e09-375f9b193048 · outbound

This paper cites task vector.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit task vector

Reference 53

Resolution
verified fuzzy
raw_fallback, observed 2026-08-10T20:15:29.269114Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

source=pdf_text observed=2026-08-10T20:15:28.472597Z digest=sha256:2338847f9f6a0a45e962cf4b6042836eed459d51ab4352b485e91f0c50fe3bfc

Observation 1611cdfc-382d-4c22-9ed8-2fcc11f1183d · outbound

This paper cites an unresolved cited work.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Unresolved cited work

Reference 55

Resolution
unresolved
raw_fallback, observed 2026-08-10T20:15:29.239255Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

source=pdf_text observed=2026-08-10T20:15:28.481781Z digest=sha256:119508d60d3fb9dbf81f15c33e16e791c61cf66445b535c5ebeb3654e7599fa6

Observation fb691456-b235-409e-b7a1-f4e5d628191a · outbound

This paper cites an unresolved cited work.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Unresolved cited work

Reference 56

Resolution
unresolved
raw_fallback, observed 2026-08-10T20:15:29.225002Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

source=pdf_text observed=2026-08-10T20:15:28.486959Z digest=sha256:29819453c61fda2def282ac81f2e16ad4d34989f7afdc3bf0f249f0f2d6a6c05

Observation 54e2455e-f931-48e0-88f5-7dd06fa6febf · outbound

This paper cites S” and 256 for notation “L.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit S” and 256 for notation “L

Reference 57

Resolution
verified fuzzy
raw_fallback, observed 2026-08-10T20:15:29.209952Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

source=pdf_text observed=2026-08-10T20:15:28.491785Z digest=sha256:1f370e9793b647bbca7d9d79b33acacec832dcd677a7b4eb760476af8d17906c

Observation 61213c96-f7a2-417e-a4a5-aae8c13f9183 · outbound

This paper cites Raventos, M.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Raventos, M

Reference 2019

Resolution
verified fuzzy
raw_fallback, observed 2026-08-10T20:15:29.330343Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

source=pdf_text observed=2026-08-10T20:15:28.393574Z digest=sha256:a03f27733507f2c61d5c5869ca89d091d03c753312275d54aff9ca14e723d4ff

Observation 72e52769-bb56-4be9-aab3-3b2270e0fd77 · outbound

This paper cites Understanding In-Context Learning in Transformers and LLMs by Learning to Learn Discrete Functions.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Understanding In-Context Learning in Transformers and LLMs by Learning to Learn Discrete Functions

Reference 2020

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.243424Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.243424Z digest=sha256:8fc7a7cb1e027487e9b63201045a24b22281136cb11f1b15efc2c2ea4a6ce248

Observation e4a3b6fa-6c40-4c47-91bd-3f607348cdc2 · outbound

This paper cites In-context learning and Occam's razor.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit In-context learning and Occam's razor

Reference 2021

Resolution
verified exact
local_arxiv, observed 2026-08-10T20:15:29.064729Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

source=pdf_text observed=2026-08-10T20:15:28.264082Z digest=sha256:142a31d8dac77d61a879c6192b13229f747efc91fa059b01236ec0b531d76b17

Observation 09e07dbc-3428-452a-b94e-e8a2a66e62dd · outbound

This paper cites In-Context Language Learning: Architectures and Algorithms.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit In-Context Language Learning: Architectures and Algorithms

Reference 2022

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.224107Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.224107Z digest=sha256:6da3a9aaf0be0dcda43cf2db92f1d9cf565845e36bc97deff5c79026102fdf3d

Observation 94295771-02f1-41d6-916d-673db3cb9df9 · outbound

This paper cites Language Models are Few-Shot Learners.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Language Models are Few-Shot Learners

Reference 2023

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.248488Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.248488Z digest=sha256:e79c87e314a6f0346ab6aa508f9be9c90470803dc5b040de717d831c7d7460b5

Observation 20c96a51-2827-41a9-a121-d9cc56ccd69d · outbound

This paper cites Transformers as Statisticians: Provable In-Context Learning with In-Context Algorithm Selection.

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit Transformers as Statisticians: Provable In-Context Learning with In-Context Algorithm Selection

Reference 2024

Resolution
unresolved
no resolver link, observed 2026-08-10T20:15:28.229076Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-10T20:15:28.229076Z digest=sha256:eea02421bcc6c3a7f68bb93cb4ab3bf0cb2e0899b1755dd23f5de5f11037e8dd

Pith citing papers

Observation c4497d80-102c-46a5-8124-e43cdbf1b82f · inbound

Adaptive Task Vectors for Large Language Models cites this paper.

Adaptive Task Vectors for Large Language Models Task Vectors in In-Context Learning: Emergence, Formation, and Benefit

Reference 13

Resolution
unresolved
no resolver link, observed 2026-08-07T11:12:59.912293Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-07T11:12:59.912293Z digest=sha256:85fee19719bfcb04e6b7fee0872827d628924e42f6221b8e119b1b76b0193d8e

Observation 6d173612-cf19-49df-9097-e44561f9a700 · inbound

Understanding Task Vectors in In-Context Learning: Emergence, Functionality, and Limitations cites this paper.

Understanding Task Vectors in In-Context Learning: Emergence, Functionality, and Limitations Task Vectors in In-Context Learning: Emergence, Formation, and Benefit

Reference 24

Resolution
unresolved
no resolver link, observed 2026-08-07T05:04:25.351904Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-07T05:04:25.351904Z digest=sha256:ef5f282c9dd704ac78d7c01f7601fc02ae51f1e0819d354b29992c8c81918c42

Observation 58e0303e-f69c-461b-8d9b-bd0b818e5ba0 · inbound

Next-Token Prediction Should be Ambiguity-Sensitive: A Meta-Learning Perspective cites this paper.

Next-Token Prediction Should be Ambiguity-Sensitive: A Meta-Learning Perspective Task Vectors in In-Context Learning: Emergence, Formation, and Benefit

Reference 31

Resolution
unresolved
no resolver link, observed 2026-08-06T23:48:42.310398Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-06T23:48:42.310398Z digest=sha256:742b48ceed3631bb90b42a0da55686db4a47c20f2ed135fa457eadf66c898273

Observation c90e2980-49f0-41bb-b018-f85a69b824d1 · inbound

Embedding-based In-Context Prompt Training for Enhancing LLMs as Text Encoders cites this paper.

Embedding-based In-Context Prompt Training for Enhancing LLMs as Text Encoders Task Vectors in In-Context Learning: Emergence, Formation, and Benefit

Reference 47

Resolution
verified exact
arxiv_id, observed 2026-05-11T16:46:15.170112Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

source=arxiv_source observed=2026-05-09T15:03:33.323423Z digest=sha256:5215bc94db8b4e4549f5379e58a41ebfe0c12790ec13af463451b4e806ba619a