Paper Citation Record · LEDGER

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse

As of 17 August 2026, this Paper Citation Record lists 66 of 66 outbound references and 2 inbound Pith citation observations for arXiv:2505.16284.

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

pith.paper-citation-record.v1
2505.16284 v1

Coverage vector

measured 66 of 66 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-07T15:11:02.432690Z

measured 68 of 68 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-16T06:30:59.297886+00:00

measured 2 of 2 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-05T21:05:00.103629Z

measured 0 of 1 external citation measurements

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

Source: pith, observed 2026-08-05T21:05:04.132846Z

Reference resolution

66 of 66 outbound references displayed

verified exact16
verified fuzzy10
unresolved40
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation ae5ca992-6ec5-4dca-b6ea-df7be148469d · outbound

This paper cites Stable Video Diffusion: Scaling Latent Video Diffusion Models to Large Datasets.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Stable Video Diffusion: Scaling Latent Video Diffusion Models to Large Datasets

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-07T15:10:57.287233Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4e35be60-8cad-42e9-a3d0-40c12a25d9bf · outbound

This paper cites On the dangers of stochastic parrots: Can langu age models be too big? In Proceedings of the 2021 ACM conference on fairness, accounta bility, and trans- parency, pages 610–623,.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse On the dangers of stochastic parrots: Can langu age models be too big? In Proceedings of the 2021 ACM conference on fairness, accounta bility, and trans- parency, pages 610–623,

Reference 5

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T15:11:08.686165Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation d13856f3-6a2a-4447-981a-e3ff59db4d3c · outbound

This paper cites Longformer: The Long-Document Transformer.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Longformer: The Long-Document Transformer

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-07T15:10:57.574963Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 397d7ef6-e300-4ac1-bb06-9320443501db · outbound

This paper cites High-Order Matching for One-Step Shortcut Diffusion Models.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse High-Order Matching for One-Step Shortcut Diffusion Models

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-07T15:10:57.772656Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7081f584-5fae-4e2d-94c3-7dd542adfeb8 · outbound

This paper cites Generating Long Sequences with Sparse Transformers.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Generating Long Sequences with Sparse Transformers

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-07T15:10:57.846217Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 37ed28c8-f28f-44b9-ad8d-4ffbce2ddeda · outbound

This paper cites Fast gradient computation for rope attention in almost line ar time.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Fast gradient computation for rope attention in almost line ar time

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-07T15:10:58.064002Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9def2f06-a17f-4cba-ad90-fbee17a36fff · outbound

This paper cites Kernel den- sity estimation through density constrained near neighbor search.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Kernel den- sity estimation through density constrained near neighbor search

Reference 13

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T15:11:08.379084Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation e1c7ada7-7a00-4393-86cc-f21411dd0f20 · outbound

This paper cites HSR-Enhanced Sparse Attention Acceleration.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse HSR-Enhanced Sparse Attention Acceleration

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-07T15:10:58.408350Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d4b62067-6cb4-4d04-85ba-d72e773e1f4e · outbound

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

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-07T15:10:58.490210Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation eadad6f9-9605-421c-a73a-39a129b31740 · outbound

This paper cites Faster Robust Tensor Power Method for Arbitrary Order.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Faster Robust Tensor Power Method for Arbitrary Order

Reference 18

Resolution: verified exact; local_arxiv, observed 2026-08-07T15:11:06.390357Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 26be70bb-8984-44cb-b130-d37eff4387f7 · outbound

This paper cites Superiori ty of softmax: Unveiling the performance edge over linear attention.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Superiori ty of softmax: Unveiling the performance edge over linear attention

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-07T15:10:58.816750Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c478d835-b6c2-4113-bfc8-5d7b9b0c8a5c · outbound

This paper cites One Step Diffusion via Shortcut Models.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse One Step Diffusion via Shortcut Models

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-07T15:10:58.928180Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 69fed7b7-7f73-48cf-8523-af0a8a9116f5 · outbound

This paper cites Sagn: semantic adaptive gr aph network for skeleton-based human action recognition.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Sagn: semantic adaptive gr aph network for skeleton-based human action recognition

Reference 21

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T15:11:08.236300Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 4effcbb0-a951-42ff-9e5a-da04e9db2bb3 · outbound

This paper cites Can You Count to Nine? A Human Evaluation Benchmark for Counting Limits in Modern Text-to-Video Models.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Can You Count to Nine? A Human Evaluation Benchmark for Counting Limits in Modern Text-to-Video Models

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-07T15:10:59.137393Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8cb376a6-b392-45e1-bb4c-431867813afe · outbound

This paper cites T2VPhysBench: A First-Principles Benchmark for Physical Consistency in Text-to-Video Generation.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse T2VPhysBench: A First-Principles Benchmark for Physical Consistency in Text-to-Video Generation

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-07T15:10:59.227370Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7b4ff729-256f-4dc6-98e9-7ba0a5ad3299 · outbound

This paper cites T2VTextBench: A Human Evaluation Benchmark for Textual Control in Video Generation Models.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse T2VTextBench: A Human Evaluation Benchmark for Textual Control in Video Generation Models

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-07T15:10:59.330611Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e89f4b95-e4dd-4266-b404-8d946c58019e · outbound

This paper cites Subquadratic Algorithms and Hardness for Attention with Any Temperature.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Subquadratic Algorithms and Hardness for Attention with Any Temperature

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-07T15:10:59.474337Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1be84ea0-f6bc-42ad-8bd9-dbb8b6ce5056 · outbound

This paper cites LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-07T15:10:59.558588Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1a39ee53-ea4e-4dbe-b892-8a20a1b03f89 · outbound

This paper cites On computational limits of ﬂowar models: Express ivity and eﬃciency.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse On computational limits of ﬂowar models: Express ivity and eﬃciency

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-07T15:10:59.690542Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7c72273d-d255-4923-9ff0-cf02b744c4b2 · outbound

This paper cites An Over-parameterized Exponential Regression.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse An Over-parameterized Exponential Regression

Reference 28

Resolution: verified exact; local_arxiv, observed 2026-08-07T15:11:05.853106Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation a8ae4233-160a-4312-9e3f-16c8a62b1ade · outbound

This paper cites Fas t quantum algorithm for attention computation.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Fas t quantum algorithm for attention computation

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-07T15:10:59.958220Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ad557723-f39f-41b5-b4f4-d268540ea384 · outbound

This paper cites Diﬀere ntially private attention computation.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Diﬀere ntially private attention computation

Reference 30

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T15:11:08.041139Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation b6920609-4c27-46fa-a82d-85e362eca3a6 · outbound

This paper cites ContraNorm: A Contrastive Learning Perspective on Oversmoothing and Beyond.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse ContraNorm: A Contrastive Learning Perspective on Oversmoothing and Beyond

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-07T15:11:00.121842Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f7a78c96-b6d3-45e2-b025-322c0481cef7 · outbound

This paper cites Generalized Probabilistic Attention Mechanism in Transformers.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Generalized Probabilistic Attention Mechanism in Transformers

Reference 32

Resolution: verified exact; local_arxiv, observed 2026-08-07T15:11:05.477298Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 3c3ed1ad-f5e9-435b-a09b-13604fc7fe22 · outbound

This paper cites Computational Limits of Low-Rank Adaptation (LoRA) Fine-Tuning for Transformer Models.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Computational Limits of Low-Rank Adaptation (LoRA) Fine-Tuning for Transformer Models

Reference 33

Resolution: verified exact; local_arxiv, observed 2026-08-07T15:11:05.244812Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 691ba6f4-69df-461b-a360-6c0198c84da1 · outbound

This paper cites On Statistical Rates and Provably Efficient Criteria of Latent Diffusion Transformers (DiTs).

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse On Statistical Rates and Provably Efficient Criteria of Latent Diffusion Transformers (DiTs)

Reference 34

Resolution: verified exact; local_arxiv, observed 2026-08-07T15:11:05.074634Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 63b4de37-4f7c-4241-87f0-0acefa71aede · outbound

This paper cites Sparse Fine-tuning for Inference Acceleration of Large Language Models.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Sparse Fine-tuning for Inference Acceleration of Large Language Models

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-07T15:11:00.461703Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 464956f2-9dab-4c33-8b65-ad6b6ac61434 · outbound

This paper cites Shortened LLaMA: Depth Pruning for Large Language Models with Comparison of Retraining Methods.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Shortened LLaMA: Depth Pruning for Large Language Models with Comparison of Retraining Methods

Reference 37

Resolution: unresolved; no resolver link, observed 2026-08-07T15:11:00.526233Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ed37a3ed-7e14-4fe9-9882-4952eb62ff98 · outbound

This paper cites On the power of preconditioning in sparse linear regression.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse On the power of preconditioning in sparse linear regression

Reference 38

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T15:11:07.955089Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 2087cf65-9e10-43ee-a267-af8e5681e1d0 · outbound

This paper cites Demonstrate-Search-Predict: Composing retrieval and language models for knowledge-intensive NLP.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Demonstrate-Search-Predict: Composing retrieval and language models for knowledge-intensive NLP

Reference 39

Resolution: unresolved; no resolver link, observed 2026-08-07T15:11:00.664900Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 83447a50-a4aa-43d9-9d23-c448ddf74ac0 · outbound

This paper cites Simulation of hypergraph algorithms with looped tr ansformers.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Simulation of hypergraph algorithms with looped tr ansformers

Reference 40

Resolution: unresolved; no resolver link, observed 2026-08-07T15:11:00.703848Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8c653ab5-95e6-40f2-9e5e-82bd315108e2 · outbound

This paper cites Exploring the frontiers of softmax: Provable optimization, applications in diﬀusion m odel, and beyond.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Exploring the frontiers of softmax: Provable optimization, applications in diﬀusion m odel, and beyond

Reference 42

Resolution: verified exact; raw_fallback, observed 2026-08-07T15:11:04.574245Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 526a9e26-0eb2-4d0d-8877-990071229060 · outbound

This paper cites A Tighter Complexity Analysis of SparseGPT.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse A Tighter Complexity Analysis of SparseGPT

Reference 43

Resolution: verified exact; local_arxiv, observed 2026-08-07T15:11:04.312562Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 30edbdec-b0be-4440-a910-3012d405961f · outbound

This paper cites Differential Privacy Mechanisms in Neural Tangent Kernel Regression.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Differential Privacy Mechanisms in Neural Tangent Kernel Regression

Reference 45

Resolution: verified exact; local_arxiv, observed 2026-08-07T15:11:04.047829Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 60c79b3b-fe08-41d0-bdc4-ba09ac61ec76 · outbound

This paper cites Diﬀerential privacy of cross- attention with provable guarantee.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Diﬀerential privacy of cross- attention with provable guarantee

Reference 46

Resolution: unresolved; no resolver link, observed 2026-08-07T15:11:01.201154Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ba38dcfe-ef43-4302-a0fc-d4154dc35afb · outbound

This paper cites Tensor attention train- ing: Provably eﬃcient learning of higher-order transforme rs.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Tensor attention train- ing: Provably eﬃcient learning of higher-order transforme rs

Reference 47

Resolution: unresolved; no resolver link, observed 2026-08-07T15:11:01.239808Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a74be657-abfe-442b-b68a-6eaf51535d38 · outbound

This paper cites The Llama 3 Herd of Models.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse The Llama 3 Herd of Models

Reference 48

Resolution: unresolved; no resolver link, observed 2026-08-07T15:11:01.298262Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 44f97945-08e5-4721-867a-522f7f925b30 · outbound

This paper cites Score-based Generative Diffusion Models for Social Recommendations.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Score-based Generative Diffusion Models for Social Recommendations

Reference 49

Resolution: verified exact; local_arxiv, observed 2026-08-07T15:11:03.578696Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 65b39b20-75e1-490c-978a-93dcf0f60872 · outbound

This paper cites Do generative video models understand physical principles?.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Do generative video models understand physical principles?

Reference 50

Resolution: unresolved; no resolver link, observed 2026-08-07T15:11:01.498098Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d8606f76-e9f0-4b7a-b3b2-c20108d802a0 · outbound

This paper cites Great power, great responsibility: Recom mendations for reducing energy for training language models.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Great power, great responsibility: Recom mendations for reducing energy for training language models

Reference 51

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T15:11:07.796116Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 343f9c74-55d3-4fe5-9512-99b86e813985 · outbound

This paper cites Miti-DETR: Object Detection based on Transformers with Mitigatory Self-Attention Convergence.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Miti-DETR: Object Detection based on Transformers with Mitigatory Self-Attention Convergence

Reference 52

Resolution: verified exact; local_arxiv, observed 2026-08-07T15:11:03.404449Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation dc4723b6-eb58-4978-9a39-59a8c1a66f23 · outbound

This paper cites Osnap: Faster numerical linear algebra algorithms via sparser subspace embeddings.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Osnap: Faster numerical linear algebra algorithms via sparser subspace embeddings

Reference 53

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T15:11:07.588867Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation e37d840e-7c8b-4656-8c25-2aac22c38233 · outbound

This paper cites GPT-4 Technical Report.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse GPT-4 Technical Report

Reference 55

Resolution: unresolved; no resolver link, observed 2026-08-07T15:11:01.774327Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1c17b35d-389a-4543-8371-b819c13259c4 · outbound

This paper cites Retentive Network: A Successor to Transformer for Large Language Models.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Retentive Network: A Successor to Transformer for Large Language Models

Reference 56

Resolution: unresolved; no resolver link, observed 2026-08-07T15:11:01.812673Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2ab26d5d-2cb1-4703-8da2-06c387b504c4 · outbound

This paper cites Numerical Pruning for Efficient Autoregressive Models.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Numerical Pruning for Efficient Autoregressive Models

Reference 57

Resolution: verified exact; local_arxiv, observed 2026-08-07T15:11:03.228296Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 46724db5-2cc8-4452-bf5a-0b194988abe5 · outbound

This paper cites Revisiting Quantum Algorithms for Linear Regressions: Quadratic Speedups without Data-Dependent Parameters.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Revisiting Quantum Algorithms for Linear Regressions: Quadratic Speedups without Data-Dependent Parameters

Reference 58

Resolution: verified exact; local_arxiv, observed 2026-08-07T15:11:03.127847Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation d5c18a48-6338-426b-9fe2-8e0adb748371 · outbound

This paper cites Transformer Dissection: A Unified Understanding of Transformer's Attention via the Lens of Kernel.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Transformer Dissection: A Unified Understanding of Transformer's Attention via the Lens of Kernel

Reference 59

Resolution: unresolved; no resolver link, observed 2026-08-07T15:11:01.988890Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c1b239f0-a21f-4541-9e4c-a89e5007a8a8 · outbound

This paper cites Alignab: Pareto- optimal energy alignment for designing nature-like antibo dies.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Alignab: Pareto- optimal energy alignment for designing nature-like antibo dies

Reference 60

Resolution: verified exact; raw_fallback, observed 2026-08-07T15:11:02.930307Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation d2582f56-ee71-44af-b8c9-9908d113dc60 · outbound

This paper cites Ev idence-aware fake news detection with graph neural networks.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Ev idence-aware fake news detection with graph neural networks

Reference 61

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T15:11:07.381558Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation bcb163b4-5db3-4f3a-b791-e068cd32876f · outbound

This paper cites Towards Better Multi-head Attention via Channel-wise Sample Permutation.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Towards Better Multi-head Attention via Channel-wise Sample Permutation

Reference 62

Resolution: verified exact; local_arxiv, observed 2026-08-07T15:11:02.647244Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 726141bb-2a40-426a-9ef6-1ff457cecb61 · outbound

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

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Trained Transformers Learn Linear Models In-Context

Reference 63

Resolution: unresolved; no resolver link, observed 2026-08-07T15:11:02.235661Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7d75fe4a-da1d-4538-ab86-86b1db4109cd · outbound

This paper cites Graph unlearning with eﬃcient partia l retraining.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Graph unlearning with eﬃcient partia l retraining

Reference 64

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T15:11:07.171546Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation bdbbbe23-e1a9-435c-a933-f82ae54b19ee · outbound

This paper cites KDEformer: Accelerating Transformers via Kernel Density Estimation.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse KDEformer: Accelerating Transformers via Kernel Density Estimation

Reference 65

Resolution: unresolved; no resolver link, observed 2026-08-07T15:11:02.385646Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation caa6a95b-2b78-41d0-b4bd-0eb87c82a535 · outbound

This paper cites LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attention.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attention

Reference 66

Resolution: unresolved; no resolver link, observed 2026-08-07T15:11:02.432690Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 03341e0c-7b97-43e3-aef5-43a475fe8e1e · outbound

This paper cites Mind the Gap: a Spectral Analysis of Rank Collapse and Signal Propagation in Attention Layers.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Mind the Gap: a Spectral Analysis of Rank Collapse and Signal Propagation in Attention Layers

Reference 2013

Resolution: unresolved; no resolver link, observed 2026-08-07T15:11:01.736227Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 02f9f4f3-7bc7-4560-968c-67a3b5d2ffa0 · outbound

This paper cites Text-to-image diﬀusion models canno t count, and prompt reﬁnement cannot help.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Text-to-image diﬀusion models canno t count, and prompt reﬁnement cannot help

Reference 2014

Resolution: unresolved; no resolver link, observed 2026-08-07T15:10:57.684027Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7810ca21-9f36-4c33-9a35-b3b10170829d · outbound

This paper cites Conv-Basis: A New Paradigm for Efficient Attention Inference and Gradient Computation in Transformers.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Conv-Basis: A New Paradigm for Efficient Attention Inference and Gradient Computation in Transformers

Reference 2016

Resolution: unresolved; no resolver link, observed 2026-08-07T15:11:00.842439Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 21c247cb-52a6-4447-9879-a090bceb86d2 · outbound

This paper cites Always Skip Attention.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Always Skip Attention

Reference 2017

Resolution: verified exact; local_arxiv, observed 2026-08-07T15:11:04.926625Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 7c0421dc-ad71-43c5-9657-25175df3b7a0 · outbound

This paper cites Streaming Kernel PCA Algorithm With Small Space.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Streaming Kernel PCA Algorithm With Small Space

Reference 2018

Resolution: verified exact; local_arxiv, observed 2026-08-07T15:11:06.573307Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation f096e649-0a24-48b1-9744-c62ca73fa541 · outbound

This paper cites Scaling Instruction-Finetuned Language Models.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Scaling Instruction-Finetuned Language Models

Reference 2019

Resolution: unresolved; no resolver link, observed 2026-08-07T15:10:57.952918Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b55a5859-5eb5-4dbc-a6fe-bcefce5b82ba · outbound

This paper cites Circuit Complexity Bounds for RoPE-based Transformer Architecture.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Circuit Complexity Bounds for RoPE-based Transformer Architecture

Reference 2020

Resolution: unresolved; no resolver link, observed 2026-08-07T15:10:58.286424Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8383b2aa-946e-441e-8191-310cea6f1728 · outbound

This paper cites Videophy: Eval- uating physical commonsense for video generation.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Videophy: Eval- uating physical commonsense for video generation

Reference 2021

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T15:11:08.542715Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 27335392-8eab-45ff-b9d4-e10df6807577 · outbound

This paper cites Multi-Layer Transformers Gradient Can be Approximated in Almost Linear Time.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Multi-Layer Transformers Gradient Can be Approximated in Almost Linear Time

Reference 2022

Resolution: unresolved; no resolver link, observed 2026-08-07T15:11:01.085476Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a591c357-6ffc-4e07-8735-366c68f1ed82 · outbound

This paper cites The geometry of BERT.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse The geometry of BERT

Reference 2023

Resolution: unresolved; no resolver link, observed 2026-08-07T15:10:57.334960Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 621ca083-c8d5-43ac-958c-942fd50c1dbb · outbound

This paper cites Fast RoPE Attention: Combining the Polynomial Method and Fast Fourier Transform.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Fast RoPE Attention: Combining the Polynomial Method and Fast Fourier Transform

Reference 2024

Resolution: unresolved; no resolver link, observed 2026-08-07T15:10:57.183708Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0b777600-3eb0-4152-a720-4b038b431ba4 · outbound

This paper cites Why do LLMs attend to the first token?.

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse Why do LLMs attend to the first token?

Reference 2025

Resolution: unresolved; no resolver link, observed 2026-08-07T15:10:57.273068Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation d2d0fa49-e60d-42c1-aa89-fb76debf5d82 · inbound

Attention's forward pass and Frank-Wolfe cites this paper.

Attention's forward pass and Frank-Wolfe Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse

Reference 2024

Resolution: verified exact; local_arxiv, observed 2026-08-05T21:05:04.193376Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation e48fb2be-2297-48ac-8c9a-88159c794113 · inbound

Transforming Rank: How Architecture Navigates the Spectral Pathologies of Depth cites this paper.

Transforming Rank: How Architecture Navigates the Spectral Pathologies of Depth Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-02T03:08:33.307962Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.