Rethinking Associative Memory Mechanism in Induction Head | Paper Citation Record

Observation ad9ad3f7-c3a7-47f2-af81-6b5cadc92815 · outbound

This paper cites Transformers learn to implement preconditioned gradient descent for in-context learning.Advances in Neural Information Pro- cessing Systems, 36:45614–45650, 2023.

Rethinking Associative Memory Mechanism in Induction Head Transformers learn to implement preconditioned gradient descent for in-context learning.Advances in Neural Information Pro- cessing Systems, 36:45614–45650, 2023

Reference 1

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T15:06:31.734005Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation 5fcfd14c-5e88-4593-b2dd-f41ce70de5e0 · outbound

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

Rethinking Associative Memory Mechanism in Induction Head What learning algorithm is in-context learning? Investigations with linear models

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.600927Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4a52f7d3-fa01-4147-86a9-9795fd03ec61 · outbound

This paper cites Learning patterns and pattern se- quences by self-organizing nets of threshold el- ements.IEEE Transactions on computers, 100 (11):1197–1206, 1972.

Rethinking Associative Memory Mechanism in Induction Head Learning patterns and pattern se- quences by self-organizing nets of threshold el- ements.IEEE Transactions on computers, 100 (11):1197–1206, 1972

Reference 3

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T15:06:31.720616Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation ab9b470f-1706-4dd2-bb51-e87a3230bc6f · outbound

This paper cites Rethinking the role of scale for in-context learning: An interpretability-based case study at 66 billion scale.

Rethinking Associative Memory Mechanism in Induction Head Rethinking the role of scale for in-context learning: An interpretability-based case study at 66 billion scale

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.609373Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 770975e3-57f9-421f-b136-dd1ee229d233 · outbound

This paper cites Birth of a transformer: A memory viewpoint.Advances in Neural Information Processing Systems, 36, 2024.

Rethinking Associative Memory Mechanism in Induction Head Birth of a transformer: A memory viewpoint.Advances in Neural Information Processing Systems, 36, 2024

Reference 5

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T15:06:31.701059Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation 99d65248-07e8-4d75-8004-032ce4e82687 · outbound

This paper cites Language Models are Few-Shot Learners.

Rethinking Associative Memory Mechanism in Induction Head Language Models are Few-Shot Learners

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.617684Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e66e1b42-dc71-4e7a-bfda-f38f5c709797 · outbound

This paper cites Scaling Laws for Associative Memories.

Rethinking Associative Memory Mechanism in Induction Head Scaling Laws for Associative Memories

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.622528Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cf83075d-3f7d-41a7-b93c-ec3785888d09 · outbound

This paper cites Learning Associative Memories with Gradient Descent.

Rethinking Associative Memory Mechanism in Induction Head Learning Associative Memories with Gradient Descent

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.627196Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 677af7ad-3f47-4c17-bdde-528c146da339 · outbound

This paper cites Distributional Associations vs In-Context Reasoning: A Study of Feed-forward and Attention Layers.

Rethinking Associative Memory Mechanism in Induction Head Distributional Associations vs In-Context Reasoning: A Study of Feed-forward and Attention Layers

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.631004Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 22a97b4d-9974-4b12-88db-2eab6e4ca4f0 · outbound

This paper cites Unveiling Induction Heads: Provable Training Dynamics and Feature Learning in Transformers.

Rethinking Associative Memory Mechanism in Induction Head Unveiling Induction Heads: Provable Training Dynamics and Feature Learning in Transformers

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.635308Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation dc22d3c5-2b82-425c-b83d-a18559879a84 · outbound

This paper cites Towards automated circuit discovery for mechanistic interpretability.Advances in Neu- ral Information Processing Systems, 36:16318– 16352, 2023.

Rethinking Associative Memory Mechanism in Induction Head Towards automated circuit discovery for mechanistic interpretability.Advances in Neu- ral Information Processing Systems, 36:16318– 16352, 2023

Reference 11

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T15:06:31.689608Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation 5788a190-bb49-418f-ad9b-f8b08acd9f79 · outbound

This paper cites Why can GPT learn in-context? language models secretly perform gradient descent as meta-optimizers.

Rethinking Associative Memory Mechanism in Induction Head Why can GPT learn in-context? language models secretly perform gradient descent as meta-optimizers

Reference 12

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T15:06:31.678314Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation 88d5cdf7-e290-4377-bb42-62f63a9b5bb2 · outbound

This paper cites Analyzing Transformers in Embedding Space.

Rethinking Associative Memory Mechanism in Induction Head Analyzing Transformers in Embedding Space

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.647483Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ec3aa380-7784-4112-a736-b452ad5c6203 · outbound

This paper cites BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding.

Rethinking Associative Memory Mechanism in Induction Head BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.651362Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 74207315-bf49-49f4-9962-8ee23518e8c7 · outbound

This paper cites The Evolution of Statistical Induction Heads: In-Context Learning Markov Chains.

Rethinking Associative Memory Mechanism in Induction Head The Evolution of Statistical Induction Heads: In-Context Learning Markov Chains

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.659771Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3b4bc33e-6a47-431c-91db-fdbc62d89c40 · outbound

This paper cites A mathematical framework for transformer circuits.Transformer Circuits Thread, 1(1):12, 2021.

Rethinking Associative Memory Mechanism in Induction Head A mathematical framework for transformer circuits.Transformer Circuits Thread, 1(1):12, 2021

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.664818Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 09d976cf-7e7d-472d-b96d-735a4829f5e3 · outbound

This paper cites Learning Transformer Programs.

Rethinking Associative Memory Mechanism in Induction Head Learning Transformer Programs

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.669296Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d9e40b71-21c9-4313-a9b3-07b867c9e0d7 · outbound

This paper cites Transformer Feed-Forward Layers Are Key-Value Memories.

Rethinking Associative Memory Mechanism in Induction Head Transformer Feed-Forward Layers Are Key-Value Memories

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.674205Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bb9fb680-f0c6-4e8e-ae0a-7ae934f22b69 · outbound

This paper cites Dissecting Recall of Factual Associations in Auto-Regressive Language Models.

Rethinking Associative Memory Mechanism in Induction Head Dissecting Recall of Factual Associations in Auto-Regressive Language Models

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.680594Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3a3d3316-69ef-487d-834a-05c6638109a9 · outbound

This paper cites DeBERTa: Decoding-enhanced BERT with Disentangled Attention.

Rethinking Associative Memory Mechanism in Induction Head DeBERTa: Decoding-enhanced BERT with Disentangled Attention

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.684936Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3de8f2f8-9f88-4f7b-98a7-b46e32c4cd01 · outbound

This paper cites Neural networks and physical systems with emergent collective computational abilities.Proceedings of the national academy of sciences, 79(8):2554–2558, 1982.

Rethinking Associative Memory Mechanism in Induction Head Neural networks and physical systems with emergent collective computational abilities.Proceedings of the national academy of sciences, 79(8):2554–2558, 1982

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.690191Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ffe3c293-9935-4769-b01f-e2a791b60762 · outbound

This paper cites Neurons with graded response have collective computational properties like those of two-state neurons.Proceedings of the national academy of sciences, 81(10):3088–3092, 1984.

Rethinking Associative Memory Mechanism in Induction Head Neurons with graded response have collective computational properties like those of two-state neurons.Proceedings of the national academy of sciences, 81(10):3088–3092, 1984

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.693846Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 405f6e6a-499c-4ef6-a2cd-c079b537731b · outbound

This paper cites In-Context Convergence of Transformers.

Rethinking Associative Memory Mechanism in Induction Head In-Context Convergence of Transformers

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.697406Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation df9221f0-c7a6-4463-adde-9735ccaa2c92 · outbound

This paper cites Improve Transformer Models with Better Relative Position Embeddings.

Rethinking Associative Memory Mechanism in Induction Head Improve Transformer Models with Better Relative Position Embeddings

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.701158Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a4705b2a-6514-4bfc-8811-c467232b32b2 · outbound

This paper cites Length Generalization in Arithmetic Transformers.

Rethinking Associative Memory Mechanism in Induction Head Length Generalization in Arithmetic Transformers

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.705650Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3d26bea1-078d-482a-b649-cb1d9a2266a6 · outbound

This paper cites Do LLMs dream of elephants (when told not to)? Latent concept association and associative memory in transformers.

Rethinking Associative Memory Mechanism in Induction Head Do LLMs dream of elephants (when told not to)? Latent concept association and associative memory in transformers

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.709589Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cd1d0fcd-1358-4580-81a4-86ad7be4aebd · outbound

This paper cites The unreasonable effec- tiveness of recurrent neural networks, May.

Rethinking Associative Memory Mechanism in Induction Head The unreasonable effec- tiveness of recurrent neural networks, May

Reference 28

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T15:06:31.646169Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation b6e6ac95-8f35-477e-b09d-116a53dd0d9f · outbound

This paper cites The impact of positional encoding on length generalization in transform- ers.Advances in Neural Information Processing Systems, 36, 2024.

Rethinking Associative Memory Mechanism in Induction Head The impact of positional encoding on length generalization in transform- ers.Advances in Neural Information Processing Systems, 36, 2024

Reference 29

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T15:06:31.621932Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation 0f47e479-a0b7-4b7b-aea8-2ac7727b117a · outbound

This paper cites Rethinking Positional Encoding in Language Pre-training.

Rethinking Associative Memory Mechanism in Induction Head Rethinking Positional Encoding in Language Pre-training

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.724755Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c9fa5e60-f296-4897-9045-37804af49ae7 · outbound

This paper cites SHAPE: Shifted Absolute Position Embedding for Transformers.

Rethinking Associative Memory Mechanism in Induction Head SHAPE: Shifted Absolute Position Embedding for Transformers

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.728751Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3fe21f49-3d5a-44de-96dd-9230712d2f4a · outbound

This paper cites Dense Associative Memory for Pattern Recognition.

Rethinking Associative Memory Mechanism in Induction Head Dense Associative Memory for Pattern Recognition

Reference 32

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.732697Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2eb14e85-5437-418a-ada0-31bbcf2758ed · outbound

This paper cites Mechan- ics of next token prediction with self-attention.

Rethinking Associative Memory Mechanism in Induction Head Mechan- ics of next token prediction with self-attention

Reference 33

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T15:06:31.610732Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation 6ca3330d-7d3b-4d85-a77c-2590a12a545b · outbound

This paper cites Cape: Encoding relative positions with continuous augmented positional embeddings.

Rethinking Associative Memory Mechanism in Induction Head Cape: Encoding relative positions with continuous augmented positional embeddings

Reference 34

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T15:06:31.598275Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation e1c6db29-71a5-4c9a-bc52-85bece87d44a · outbound

This paper cites an unresolved cited work.

Rethinking Associative Memory Mechanism in Induction Head Unresolved cited work

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.743989Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9790ab9f-e076-48f5-b968-a56a684d3f6e · outbound

This paper cites One Step of Gradient Descent is Provably the Optimal In-Context Learner with One Layer of Linear Self-Attention.

Rethinking Associative Memory Mechanism in Induction Head One Step of Gradient Descent is Provably the Optimal In-Context Learner with One Layer of Linear Self-Attention

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.747797Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 408ee123-c242-4a52-a9b3-1e2b60866e00 · outbound

This paper cites Locating and editing factual associations in gpt.Advances in Neural Informa- tion Processing Systems, 35:17359–17372, 2022.

Rethinking Associative Memory Mechanism in Induction Head Locating and editing factual associations in gpt.Advances in Neural Informa- tion Processing Systems, 35:17359–17372, 2022

Reference 37

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T15:06:31.585732Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation 7bbce405-afcc-456d-8392-755ec5ee8ae4 · outbound

This paper cites Efficient Estimation of Word Representations in Vector Space.

Rethinking Associative Memory Mechanism in Induction Head Efficient Estimation of Word Representations in Vector Space

Reference 38

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.755450Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation eebd06bf-62ef-40a1-8909-8e495c885af2 · outbound

This paper cites Progress measures for grokking via mechanistic interpretability.

Rethinking Associative Memory Mechanism in Induction Head Progress measures for grokking via mechanistic interpretability

Reference 39

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.758643Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c0c277ba-e470-4974-894c-25ead4f3af73 · outbound

This paper cites On the rela- tion between position information and sentence length in neural machine translation.

Rethinking Associative Memory Mechanism in Induction Head On the rela- tion between position information and sentence length in neural machine translation

Reference 40

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T15:06:31.573510Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation 8c183ed2-280b-45ad-925c-ea1580ad48cc · outbound

This paper cites How Transformers Learn Causal Structure with Gradient Descent.

Rethinking Associative Memory Mechanism in Induction Head How Transformers Learn Causal Structure with Gradient Descent

Reference 41

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.765076Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8a28eed7-1b16-4efe-add8-acee5a370ced · outbound

This paper cites In-context Learning and Induction Heads.

Rethinking Associative Memory Mechanism in Induction Head In-context Learning and Induction Heads

Reference 42

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.768808Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4808d3a3-547b-49ea-ae2b-25840acaf06e · outbound

This paper cites Train Short, Test Long: Attention with Linear Biases Enables Input Length Extrapolation.

Rethinking Associative Memory Mechanism in Induction Head Train Short, Test Long: Attention with Linear Biases Enables Input Length Extrapolation

Reference 43

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.772493Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9855d331-a269-4b5f-9e43-7947f80d47a9 · outbound

This paper cites Improving language understanding by generative pre-training.

Rethinking Associative Memory Mechanism in Induction Head Improving language understanding by generative pre-training

Reference 44

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.776578Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a881571e-cfd5-4788-a5c4-ce3a1064676b · outbound

This paper cites Lan- guage models are unsupervised multitask learners.

Rethinking Associative Memory Mechanism in Induction Head Lan- guage models are unsupervised multitask learners

Reference 45

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.780438Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 04440c7b-a158-497b-929e-384e98a2d0fd · outbound

This paper cites Exploring the limits of transfer learning with a unified text- to-text transformer.Journal of machine learning research, 21(140):1–67, 2020.

Rethinking Associative Memory Mechanism in Induction Head Exploring the limits of transfer learning with a unified text- to-text transformer.Journal of machine learning research, 21(140):1–67, 2020

Reference 46

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T15:06:31.544560Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation 0c20aa39-d29b-4089-b353-2fccbbae5a08 · outbound

This paper cites Identifying semantic induction heads to under- stand in-context learning.

Rethinking Associative Memory Mechanism in Induction Head Identifying semantic induction heads to under- stand in-context learning

Reference 47

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.787727Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 31744b4e-9f9e-4071-8aa9-78c0fb6209bb · outbound

This paper cites Randomized Positional Encodings Boost Length Generalization of Transformers.

Rethinking Associative Memory Mechanism in Induction Head Randomized Positional Encodings Boost Length Generalization of Transformers

Reference 48

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.791311Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7f2acdbf-7db9-4291-b069-568d3d77749a · outbound

This paper cites Self-Attention with Relative Position Representations.

Rethinking Associative Memory Mechanism in Induction Head Self-Attention with Relative Position Representations

Reference 49

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.795596Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 18acadc7-4960-4b92-bdf3-a8320901d0a5 · outbound

This paper cites The Curious Case of Absolute Position Embeddings.

Rethinking Associative Memory Mechanism in Induction Head The Curious Case of Absolute Position Embeddings

Reference 50

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.799756Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1f3b8b56-3887-41d1-9bea-a43fec2a5e8d · outbound

This paper cites Roformer: En- hanced transformer with rotary position embed- ding.Neurocomputing, 568:127063, 2024.

Rethinking Associative Memory Mechanism in Induction Head Roformer: En- hanced transformer with rotary position embed- ding.Neurocomputing, 568:127063, 2024

Reference 51

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T15:06:31.530811Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation 8aca132c-1795-45d4-b7c4-2aabe5584f63 · outbound

This paper cites Attention is all you need.Advances in Neural Information Processing Systems, 2017.

Rethinking Associative Memory Mechanism in Induction Head Attention is all you need.Advances in Neural Information Processing Systems, 2017

Reference 52

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.807818Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e5010167-3a79-43ba-a5ee-c011e867700a · outbound

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

Rethinking Associative Memory Mechanism in Induction Head Transformers learn in-context by gradient descent

Reference 53

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T15:06:31.511354Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation b4d96e97-2245-4016-b195-26912d364830 · outbound

This paper cites Encoding word order in complex embeddings.

Rethinking Associative Memory Mechanism in Induction Head Encoding word order in complex embeddings

Reference 54

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.815527Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8f81d085-72d5-4116-997a-f59db5b94876 · outbound

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

Rethinking Associative Memory Mechanism in Induction Head Interpretability in the Wild: a Circuit for Indirect Object Identification in GPT-2 small

Reference 55

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.819522Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 304d6006-80e8-45bc-a4a1-62d42ef619ca · outbound

This paper cites La- bel words are anchors: An information flow per- spective for understanding in-context learning.

Rethinking Associative Memory Mechanism in Induction Head La- bel words are anchors: An information flow per- spective for understanding in-context learning

Reference 56

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.823576Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 02dba459-3199-4f25-bc32-d46104955612 · outbound

This paper cites The learnability of in-context learning.Advances in Neural Information Processing Systems, 36, 2024.

Rethinking Associative Memory Mechanism in Induction Head The learnability of in-context learning.Advances in Neural Information Processing Systems, 36, 2024

Reference 57

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T15:06:31.496953Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation e8d6887f-34fe-45fa-9531-5f89a01992c9 · outbound

This paper cites Self-adaptive in-context learn- ing: An information compression perspective for in-context example selection and ordering.

Rethinking Associative Memory Mechanism in Induction Head Self-adaptive in-context learn- ing: An information compression perspective for in-context example selection and ordering

Reference 58

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.831223Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9de2dec0-f3f6-40be-8489-1f6536867ffb · outbound

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

Rethinking Associative Memory Mechanism in Induction Head An Explanation of In-context Learning as Implicit Bayesian Inference

Reference 59

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.834466Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9ff11a43-7c61-4cfd-b0f2-e7778c8cc7b6 · outbound

This paper cites Memory Mosaics.

Rethinking Associative Memory Mechanism in Induction Head Memory Mosaics

Reference 60

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.838044Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4f26a7b3-ed9d-47fb-9a2c-994c7c5cd7c5 · outbound

This paper cites Trained Transformers Learn Linear Models In-Context.

Rethinking Associative Memory Mechanism in Induction Head Trained Transformers Learn Linear Models In-Context

Reference 61

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.841729Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 982f6daf-278c-4cbf-8cd8-7aa6197c76a5 · outbound

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

Rethinking Associative Memory Mechanism in Induction Head What and How does In-Context Learning Learn? Bayesian Model Averaging, Parameterization, and Generalization

Reference 62

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.845577Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 70e3f72b-bc88-4bf9-bbe0-2a55a11bcb12 · outbound

This paper cites Length Extrapolation of Transformers: A Survey from the Perspective of Positional Encoding.

Rethinking Associative Memory Mechanism in Induction Head Length Extrapolation of Transformers: A Survey from the Perspective of Positional Encoding

Reference 63

Resolution: unresolved; no resolver link, observed 2026-08-11T15:06:30.849934Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 13cad500-a210-4cb6-9155-40a5e92668f1 · outbound

This paper cites In other words, let tq be the first occurrence position.

Rethinking Associative Memory Mechanism in Induction Head In other words, let tq be the first occurrence position

Reference 65

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T15:06:31.483495Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation 62f889b7-2970-4b61-aa7b-dde41e405e2c · outbound

This paper cites an unresolved cited work.

Rethinking Associative Memory Mechanism in Induction Head Unresolved cited work

Reference 66

Resolution: unresolved; raw_fallback, observed 2026-08-11T15:06:31.471809Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation 24c3e12b-38d5-4cc4-94a7-8eac6bf452ed · outbound

This paper cites an unresolved cited work.

Rethinking Associative Memory Mechanism in Induction Head Unresolved cited work

Reference 67

Resolution: unresolved; raw_fallback, observed 2026-08-11T15:06:31.460118Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation 35477c55-6548-4248-8d4e-e5908bd88f33 · outbound

This paper cites We employ zero- initialization and carry out a single gradient de- scent step.

Rethinking Associative Memory Mechanism in Induction Head We employ zero- initialization and carry out a single gradient de- scent step

Reference 68

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T15:06:31.447744Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation ebab3649-9b11-40f5-af2f-d9d2e0555eca · outbound

This paper cites TX t=to W 2 V 1 t Φ1wE(zto)|y=k # − 1 T ·E.

Rethinking Associative Memory Mechanism in Induction Head TX t=to W 2 V 1 t Φ1wE(zto)|y=k # − 1 T ·E

Reference 69

Resolution: malformed identifier; raw_fallback, observed 2026-08-11T15:06:31.433540Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation cff2710d-1b3c-4f6f-94ff-2d1ee0eec58d · outbound

This paper cites an unresolved cited work.

Rethinking Associative Memory Mechanism in Induction Head Unresolved cited work

Reference 70

Resolution: unresolved; raw_fallback, observed 2026-08-11T15:06:31.420465Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation 74b5f145-c93c-451b-a569-bf7c8a2d5ede · outbound

This paper cites an unresolved cited work.

Rethinking Associative Memory Mechanism in Induction Head Unresolved cited work

Reference 71

Resolution: unresolved; raw_fallback, observed 2026-08-11T15:06:31.409305Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation d05a34f2-8366-4b55-985b-a50096882542 · outbound

This paper cites an unresolved cited work.

Rethinking Associative Memory Mechanism in Induction Head Unresolved cited work

Reference 72

Resolution: unresolved; raw_fallback, observed 2026-08-11T15:06:31.397317Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation 44ec3885-ad92-48bb-ae04-dbfea01fa909 · outbound

This paper cites Everytime we generate an input sequence, we sam- ple5trigger tokens from the set of all source words, and uniformly selects the corresponding output tokens from all vocabulary.

Rethinking Associative Memory Mechanism in Induction Head Everytime we generate an input sequence, we sam- ple5trigger tokens from the set of all source words, and uniformly selects the corresponding output tokens from all vocabulary

Reference 73

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T15:06:31.384668Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation 6ca3bc0c-e140-4937-9ae8-2bb75487a513 · outbound

This paper cites Accessed: 2024-12-15.

Rethinking Associative Memory Mechanism in Induction Head Accessed: 2024-12-15

Reference 2015

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T15:06:31.634491Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Coverage vector

Reference resolution

Outbound references

Pith citing papers