Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing | Paper Citation Record

Observation 9a850f0d-6b5d-4324-a5a0-8e0e3a606957 · outbound

This paper cites Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:51.657826Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 90576262-5da8-492d-ae7d-cb02bda79628 · outbound

This paper cites GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:51.939144Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4206ba9c-d23f-4ae0-84e6-b3129ca4fe83 · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:52.099338Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5da48126-9546-414f-88a9-206877dac43f · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:52.253509Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 86a27565-70d4-4762-980d-06c74fe6e9c8 · outbound

This paper cites 2019.{CSI}{ NN}: Reverse engineering of neural network architectures through electromagnetic side channel.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing 2019.{CSI}{ NN}: Reverse engineering of neural network architectures through electromagnetic side channel

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:52.382521Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 529d8384-7001-4f9c-b503-3d4eb6bbac29 · outbound

This paper cites Speculative Streaming: Fast LLM Inference without Auxiliary Models.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Speculative Streaming: Fast LLM Inference without Auxiliary Models

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:52.548132Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation caccdfe6-ebbb-43af-90b2-dea6b9caa215 · outbound

This paper cites Lee, Dem- ing Chen, and Tri Dao.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Lee, Dem- ing Chen, and Tri Dao

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:52.719770Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 959e44d2-7aa2-4c58-ad8d-8439cf284937 · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:52.857429Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 81399209-2cc7-4d66-bced-21b68dbbf3c0 · outbound

This paper cites Remote Timing Attacks on Efficient Language Model Inference.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Remote Timing Attacks on Efficient Language Model Inference

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:52.980588Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f0b31db0-06a1-4bfd-88c5-b84f210d869a · outbound

This paper cites Feder Cooper, Katherine Lee, Matthew Jagielski, Milad Nasr, Arthur Conmy, Eric Wallace, David Rolnick, and Florian Tramèr.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Feder Cooper, Katherine Lee, Matthew Jagielski, Milad Nasr, Arthur Conmy, Eric Wallace, David Rolnick, and Florian Tramèr

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:53.101665Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c94d5e88-c8a5-4908-b8f6-0da4127d6f8b · outbound

This paper cites Accelerating Large Language Model Decoding with Speculative Sampling.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Accelerating Large Language Model Decoding with Speculative Sampling

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:53.229418Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ed3531fd-084b-497b-be3b-30f4b3ce0f2c · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:53.373190Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c91311fc-c99d-4751-b233-5b0f3821ab70 · outbound

This paper cites Cascade Speculative Drafting for Even Faster LLM Inference.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Cascade Speculative Drafting for Even Faster LLM Inference

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:53.518187Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5e5a8578-0338-4143-810d-e77a5a1a332a · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:53.715064Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2aa741b5-845d-4fc4-9cf7-9420d564c748 · outbound

This paper cites Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:53.820456Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5660117a-f4f3-4bca-8bc9-9447207911ca · outbound

This paper cites FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:53.947348Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7756bed3-e220-4327-b7ea-ba7567e79dba · outbound

This paper cites FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:54.079800Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ca44effc-5c2f-402e-98c7-b07bf6029e39 · outbound

This paper cites China is going to win the AI race.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing China is going to win the AI race

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:54.255995Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f5262c9f-c3fd-4c4a-a79f-97b6d02f625a · outbound

This paper cites Stealing Neural Networks via Timing Side Channels.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Stealing Neural Networks via Timing Side Channels

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:54.400330Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fc54055f-99b7-42a7-bbb9-077cd289777e · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:54.528112Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2fa0c712-650e-4afd-a248-f4a0e3da898a · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:54.709071Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8418a902-9967-40c0-96a9-5d46519ab964 · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:54.810400Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 16078283-4456-4dce-a4a4-b46f06e3008d · outbound

This paper cites I Know What You Said: Unveiling Hardware Cache Side-Channels in Local Large Language Model Inference.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing I Know What You Said: Unveiling Hardware Cache Side-Channels in Local Large Language Model Inference

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:54.914643Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 60d2bb7b-f698-49bd-903b-a3a2df515234 · outbound

This paper cites Auditing Prompt Caching in Language Model APIs.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Auditing Prompt Caching in Language Model APIs

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:54.993015Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4ecd3061-f843-448b-a88c-ef5796540ebd · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:55.086637Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 493cec01-8afe-4a40-b611-2ad94cab703f · outbound

This paper cites Hennessy and David A.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Hennessy and David A

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:55.185773Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bfe93aff-46b4-47cd-8c88-fec044d4e161 · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:55.255147Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9625c599-fe28-424b-aa55-1246f088a5e6 · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:55.325493Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e63cffaf-8873-491d-9031-3321d8d81d22 · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:55.390685Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b4ab9a25-7e28-408b-882b-5c3fb2600f57 · outbound

This paper cites Aamodt, and Timothy G.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Aamodt, and Timothy G

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:55.457039Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1f63ef2a-4df9-43bf-8289-205ecace50e3 · outbound

This paper cites Mahoney, Amir Gholami, and Kurt Keutzer.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Mahoney, Amir Gholami, and Kurt Keutzer

Reference 32

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:55.609339Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 92e71f74-c143-40ff-ad35-312e13484678 · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 33

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:55.683338Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d5cd4850-53ea-4983-9d72-13b6d1b5c61f · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 34

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:55.799096Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ca0dc78b-6206-469c-ab36-7cd3ce7d080b · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 35

Resolution: malformed identifier; no resolver link, observed 2026-08-01T09:35:55.966381Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 00979975-ca78-4a01-bb7c-dcbf22fc1f10 · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:56.167330Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fd4cc1e6-8e17-4f6b-af90-ae1c13bcc145 · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 37

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:56.307522Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3abca5a6-1c3a-4c81-8597-e13bafc96597 · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 38

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:56.446008Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 122d0a5f-1620-4849-a34a-e6de1496ce01 · outbound

This paper cites Andes: Defining and Enhancing Quality-of-Experience in LLM-Based Text Streaming Services.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Andes: Defining and Enhancing Quality-of-Experience in LLM-Based Text Streaming Services

Reference 39

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:56.582200Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f2f81839-e2ea-47da-b6e9-50a1e88e4aa8 · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 40

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:56.749480Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4b11976e-c228-42ad-94ea-ad73dc682418 · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 41

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:56.921454Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7a851fe9-5cfb-46ad-8d70-6cf2f5476036 · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 42

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:57.107524Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d1c52ddb-5254-4f62-9903-676e74e36709 · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 43

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:57.293949Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6c106b63-7709-415b-a7c1-89fd43715727 · outbound

This paper cites gpt-oss-120b & gpt-oss-20b Model Card.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing gpt-oss-120b & gpt-oss-20b Model Card

Reference 44

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:57.425554Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9bb30132-fd9c-445b-a888-2db9ac401b05 · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 45

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:57.484490Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3e0c4e50-8999-48f0-9df5-1c39d60135f3 · outbound

This paper cites Forecasting LLM Inference Performance via Hardware-Agnostic Analytical Modeling.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Forecasting LLM Inference Performance via Hardware-Agnostic Analytical Modeling

Reference 46

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:57.546394Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e0e85464-7a77-4a40-bf45-d7f3b7e59c90 · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 47

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:57.630587Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2d515a64-b1ff-4eb8-8979-691237c346cc · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 48

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:57.743800Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a6b588f0-b2ae-46bc-807f-7425f7b9e5a9 · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 49

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:57.978477Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation af55fa44-6997-499a-b6e0-a7385b1fe027 · outbound

This paper cites Mixture-of-Depths: Dynamically allocating compute in transformer-based language models.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Mixture-of-Depths: Dynamically allocating compute in transformer-based language models

Reference 50

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:58.191128Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 96db6c13-9e05-46ea-a3e8-60c57666e6b1 · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 51

Resolution: malformed identifier; no resolver link, observed 2026-08-01T09:35:58.334823Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation be945d9a-4423-435c-b2ff-10048d38ea18 · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 52

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:58.536705Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 81f670c5-52a6-4783-9d9c-0b64231b5338 · outbound

This paper cites Accelerating LLM Inference with Staged Speculative Decoding.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Accelerating LLM Inference with Staged Speculative Decoding

Reference 53

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:58.757157Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 65acffae-386b-4f66-8dd5-47d09d239f58 · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 54

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:58.959985Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ecbaad3e-54c8-4922-9f29-972ce826c981 · outbound

This paper cites SpecTr: Fast Speculative Decoding via Optimal Transport.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing SpecTr: Fast Speculative Decoding via Optimal Transport

Reference 55

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:59.177971Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 79777f1a-980e-4110-9fb7-b0afb0784f41 · outbound

This paper cites Gemini: A Family of Highly Capable Multimodal Models.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Gemini: A Family of Highly Capable Multimodal Models

Reference 56

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:59.321038Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7d076487-1ebf-4714-9463-e13a539c492a · outbound

This paper cites Gemma 2: Improving Open Language Models at a Practical Size.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Gemma 2: Improving Open Language Models at a Practical Size

Reference 57

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:59.503236Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c12dd328-df24-484d-9a30-e835be02c64a · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 58

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:59.662075Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f6969907-6bea-40ea-8d87-24b7e14bd210 · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 59

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:59.858754Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ed83be52-15c0-4c81-9ef4-de9465f0e25f · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 60

Resolution: unresolved; no resolver link, observed 2026-08-01T09:36:00.051308Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7fef8d14-3bdd-4f78-af32-78aca4eb4c8e · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 61

Resolution: unresolved; no resolver link, observed 2026-08-01T09:36:00.182766Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a0a8922e-b13d-43bf-9e4d-e0f92ebd20a4 · outbound

This paper cites LLaMA: Open and Efficient Foundation Language Models.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing LLaMA: Open and Efficient Foundation Language Models

Reference 62

Resolution: unresolved; no resolver link, observed 2026-08-01T09:36:00.328905Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9f250abb-979c-412c-9db1-da1a2f00c723 · outbound

This paper cites Gomez, Łukasz Kaiser, and Illia Polosukhin.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Gomez, Łukasz Kaiser, and Illia Polosukhin

Reference 63

Resolution: unresolved; no resolver link, observed 2026-08-01T09:36:00.458810Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c0b3a1e2-5971-4348-b7d3-080d492ef3a2 · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 64

Resolution: unresolved; no resolver link, observed 2026-08-01T09:36:00.583494Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 86a25ce0-3a0d-4889-ad95-bb1d59c46e48 · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 65

Resolution: unresolved; no resolver link, observed 2026-08-01T09:36:00.701473Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fde2ef62-2480-4da0-9ffc-a772a9a9f439 · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 66

Resolution: unresolved; no resolver link, observed 2026-08-01T09:36:00.845776Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a6e9bd11-2104-4f4f-94fb-1da759ef99d5 · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 67

Resolution: unresolved; no resolver link, observed 2026-08-01T09:36:01.028109Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9ab8f538-5b06-4951-bd86-69dc7c11f7c6 · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 68

Resolution: unresolved; no resolver link, observed 2026-08-01T09:36:01.173969Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 646424dc-e609-4914-80a2-4357ff1e9ced · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 69

Resolution: unresolved; no resolver link, observed 2026-08-01T09:36:01.470511Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3adaf89b-47ff-4be6-8640-3ada7e01fb57 · outbound

This paper cites an unresolved cited work.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Unresolved cited work

Reference 70

Resolution: unresolved; no resolver link, observed 2026-08-01T09:36:01.600968Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 02df7adc-7924-4ee8-a39d-0d0ac1fb0986 · outbound

This paper cites Fletcher, and Josep Torrellas.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Fletcher, and Josep Torrellas

Reference 71

Resolution: unresolved; no resolver link, observed 2026-08-01T09:36:01.717180Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6cb3355c-6dce-469a-b6d4-dbe26b9ef61d · outbound

This paper cites LongSpec: Long-Context Lossless Speculative Decoding with Efficient Drafting and Verification.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing LongSpec: Long-Context Lossless Speculative Decoding with Efficient Drafting and Verification

Reference 72

Resolution: unresolved; no resolver link, observed 2026-08-01T09:36:01.890755Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d4fe5736-c98f-447a-86bd-c0fc3b0f9a93 · outbound

This paper cites Time Will Tell: Timing Side Channels via Output Token Count in Large Language Models.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing Time Will Tell: Timing Side Channels via Output Token Count in Large Language Models

Reference 73

Resolution: unresolved; no resolver link, observed 2026-08-01T09:36:02.050940Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1161d847-04c4-4f65-a65e-089ade15f53a · outbound

This paper cites InputSnatch: Stealing Input in LLM Services via Timing Side-Channel Attacks.

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing InputSnatch: Stealing Input in LLM Services via Timing Side-Channel Attacks

Reference 74

Resolution: unresolved; no resolver link, observed 2026-08-01T09:36:02.185670Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ce416bb6-b089-49ee-b22d-e049e429778a · outbound

This paper cites InProceedings of the ACM/IEEE 47th Annual International Sympo- sium on Computer Architecture(Virtual Event)(ISCA ’20).

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing InProceedings of the ACM/IEEE 47th Annual International Sympo- sium on Computer Architecture(Virtual Event)(ISCA ’20)

Reference 2020

Resolution: unresolved; no resolver link, observed 2026-08-01T09:35:55.545625Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 65399667-8858-4c28-9634-89d8b236d803 · outbound

This paper cites InFindings of the Association for Computational Linguistics: EMNLP 2023, Houda Bouamor, Juan Pino, and Kalika Bali (Eds.).

Leaky Language Models: Stealing Architecture and Inference Optimizations via Per-Token Timing InFindings of the Association for Computational Linguistics: EMNLP 2023, Houda Bouamor, Juan Pino, and Kalika Bali (Eds.)

Reference 2023

Resolution: malformed identifier; no resolver link, observed 2026-08-01T09:36:01.324744Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Coverage vector

Reference resolution

Outbound references

Pith citing papers