Paper Citation Record · LEDGER

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation

As of 9 August 2026, this Paper Citation Record lists 59 of 59 outbound references and 3 inbound Pith citation observations for arXiv:2502.02789.

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

pith.paper-citation-record.v1
2502.02789 v2

Coverage vector

measured 59 of 59 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-09T11:11:17.839813Z

measured 62 of 62 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-09T06:31:02.800959+00:00

measured 3 of 3 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-07T12:39:05.263720Z

measured 0 of 1 external citation measurements

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

Source: arxiv_reference, observed 2026-07-03T14:48:33.364894Z

Reference resolution

59 of 59 outbound references displayed

verified exact0
verified fuzzy2
unresolved57
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 825cc507-4d35-4288-a66c-0e4333a50bbd · outbound

This paper cites write newline.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation write newline

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.602177Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c35c8163-c947-43dc-ac37-4657bc64e693 · outbound

This paper cites Program Synthesis with Large Language Models.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation Program Synthesis with Large Language Models

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.607441Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f0bb1da9-e101-4c37-a710-7353aed9f5c4 · outbound

This paper cites L ong B ench: A bilingual, multitask benchmark for long context understanding.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation L ong B ench: A bilingual, multitask benchmark for long context understanding

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.611991Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8ba76ba8-24d4-46dd-9062-7b6f7dbcd9c7 · outbound

This paper cites Longformer: The Long-Document Transformer.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation Longformer: The Long-Document Transformer

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.615898Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 40efdeaf-b4f3-4037-a16a-495ea9432db2 · outbound

This paper cites Sparks of Artificial General Intelligence: Early experiments with GPT-4.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation Sparks of Artificial General Intelligence: Early experiments with GPT-4

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.620625Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 233f7931-9bf8-4029-9d07-a205c366d6e2 · outbound

This paper cites Evaluating Large Language Models Trained on Code.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation Evaluating Large Language Models Trained on Code

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.624690Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 67435126-6772-4b69-b780-3fc8020cd76b · outbound

This paper cites Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.628946Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 25149e5d-9d65-43a2-9266-6357f6084a6d · outbound

This paper cites Training verifiers to solve math word problems, 2021.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation Training verifiers to solve math word problems, 2021

Reference 8

Resolution: verified fuzzy; raw_fallback, observed 2026-08-09T11:11:31.010258Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation db173266-a6eb-45d5-9fe3-470492b91798 · outbound

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

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.636804Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 82b64e1e-fb99-41bc-9b80-58758ec30ea8 · outbound

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

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.640537Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation da49d82b-1f3f-4462-a0a1-d40ea8354f05 · outbound

This paper cites DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.644345Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c793fdf5-84a3-4a65-8b54-3bb3a0111926 · outbound

This paper cites Layerskip: Enabling early exit inference and self-speculative decoding.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation Layerskip: Enabling early exit inference and self-speculative decoding

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.648359Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1afb1e5c-3e4f-4a51-9223-9c80ed6df62e · outbound

This paper cites How Far Are We From AGI: Are LLMs All We Need?.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation How Far Are We From AGI: Are LLMs All We Need?

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.652509Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0a4600c3-a957-4af5-8d74-fe913fb18020 · outbound

This paper cites A framework for few-shot language model evaluation, 07 2024 a.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation A framework for few-shot language model evaluation, 07 2024 a

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.656292Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8cc5a2cf-5255-4c68-91c9-bd138ecd0c98 · outbound

This paper cites Retrieval-Augmented Generation for Large Language Models: A Survey.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation Retrieval-Augmented Generation for Large Language Models: A Survey

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.659971Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation db3c62e1-e760-4347-9604-a1e22c7a2580 · outbound

This paper cites The Llama 3 Herd of Models.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation The Llama 3 Herd of Models

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.664015Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f41cc89b-f9d8-47f3-8744-968c435497e9 · outbound

This paper cites Measuring Massive Multitask Language Understanding.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation Measuring Massive Multitask Language Understanding

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.669337Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b1d47dfe-223c-4c48-a082-8e450037c5a3 · outbound

This paper cites KV Prediction for Improved Time to First Token.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation KV Prediction for Improved Time to First Token

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.673384Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 37a9dd95-f2aa-43f0-a172-06288745b923 · outbound

This paper cites RULER: What's the Real Context Size of Your Long-Context Language Models?.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation RULER: What's the Real Context Size of Your Long-Context Language Models?

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.677430Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6ee53d13-3743-46d1-9a53-d46014661c0b · outbound

This paper cites Mistral 7B.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation Mistral 7B

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.681510Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c0805d18-4818-4e42-8e52-cf6f52f50a36 · outbound

This paper cites LLMLingua: Compressing Prompts for Accelerated Inference of Large Language Models.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation LLMLingua: Compressing Prompts for Accelerated Inference of Large Language Models

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.685933Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 81687733-b76f-4519-b52d-389e12613747 · outbound

This paper cites MInference 1.0: Accelerating Pre-filling for Long-Context LLMs via Dynamic Sparse Attention.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation MInference 1.0: Accelerating Pre-filling for Long-Context LLMs via Dynamic Sparse Attention

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.689930Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e98294fb-c605-48be-9e27-8efe5c153e0c · outbound

This paper cites LongLLMLingua: Accelerating and Enhancing LLMs in Long Context Scenarios via Prompt Compression.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation LongLLMLingua: Accelerating and Enhancing LLMs in Long Context Scenarios via Prompt Compression

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.694065Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 40f0d174-2e08-400d-aa92-67080801e89e · outbound

This paper cites H., Gonzalez, J.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation H., Gonzalez, J

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.698253Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 730aa723-d591-417c-832f-5400fda1cef9 · outbound

This paper cites FlexPrefill: A Context-Aware Sparse Attention Mechanism for Efficient Long-Sequence Inference.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation FlexPrefill: A Context-Aware Sparse Attention Mechanism for Efficient Long-Sequence Inference

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.702371Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e97223e2-fc57-465d-8165-a03c84f75dbd · outbound

This paper cites InfiniteHiP: Extending Language Model Context Up to 3 Million Tokens on a Single GPU.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation InfiniteHiP: Extending Language Model Context Up to 3 Million Tokens on a Single GPU

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.706860Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 67b9bca9-d6bf-4eb9-834a-6212459ce350 · outbound

This paper cites Fast Inference from Transformers via Speculative Decoding.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation Fast Inference from Transformers via Speculative Decoding

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.711098Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 02d31b63-ea18-4cd1-9788-c6119ad433e5 · outbound

This paper cites Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.714944Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c1aa0ade-cbc0-4d9f-8a4d-5969cbca6a80 · outbound

This paper cites Compressing Context to Enhance Inference Efficiency of Large Language Models.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation Compressing Context to Enhance Inference Efficiency of Large Language Models

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.718608Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 73f6391c-4166-407d-973e-5850a203fe1f · outbound

This paper cites SCBench: A KV Cache-Centric Analysis of Long-Context Methods.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation SCBench: A KV Cache-Centric Analysis of Long-Context Methods

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.722512Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5662039a-e072-439e-866d-c89b879a0ff1 · outbound

This paper cites Retrieval Augmented Generation or Long-Context LLMs? A Comprehensive Study and Hybrid Approach.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation Retrieval Augmented Generation or Long-Context LLMs? A Comprehensive Study and Hybrid Approach

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.726660Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2df015bd-3b85-4f0d-bc87-472d78c5de49 · outbound

This paper cites AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration

Reference 32

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.730999Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 45bce6ca-a1cb-4663-b39f-64869f22f964 · outbound

This paper cites S., Wang, Y., and Zhang, L.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation S., Wang, Y., and Zhang, L

Reference 33

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.735194Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 12d1fef0-7650-4007-95a7-1d84b3e70cb5 · outbound

This paper cites CacheGen: KV Cache Compression and Streaming for Fast Large Language Model Serving.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation CacheGen: KV Cache Compression and Streaming for Fast Large Language Model Serving

Reference 34

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.739768Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4476ab2b-8dbc-4bc6-bb4e-ff108b27ed32 · outbound

This paper cites NLTK: The Natural Language Toolkit.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation NLTK: The Natural Language Toolkit

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.743641Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation be5abe06-8e50-4a74-b148-10464d34f148 · outbound

This paper cites CritiPrefill: A Segment-wise Criticality-based Approach for Prefilling Acceleration in LLMs.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation CritiPrefill: A Segment-wise Criticality-based Approach for Prefilling Acceleration in LLMs

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.748351Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 56fc0fe0-bdde-48b2-82e9-b1931cd5c1a5 · outbound

This paper cites Towards Efficient Generative Large Language Model Serving: A Survey from Algorithms to Systems.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation Towards Efficient Generative Large Language Model Serving: A Survey from Algorithms to Systems

Reference 37

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.752885Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 229521fa-32a6-4c05-a7a2-1ccd4e4db97d · outbound

This paper cites GPT-4 Technical Report.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation GPT-4 Technical Report

Reference 38

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.756918Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 42e8049c-d875-46f0-8f10-659c77c0504c · outbound

This paper cites SwiftKV: Fast Prefill-Optimized Inference with Knowledge-Preserving Model Transformation.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation SwiftKV: Fast Prefill-Optimized Inference with Knowledge-Preserving Model Transformation

Reference 39

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.760837Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 02c128c2-b44d-4c5a-a9b5-205f2c17f685 · outbound

This paper cites Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks

Reference 40

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.764820Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e2d7906d-5500-4d0d-aca2-5cfcc5490e8e · outbound

This paper cites L., Stickland, A.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation L., Stickland, A

Reference 41

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.768911Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0e6dc874-1a7f-4625-9aea-9f54564b78d3 · outbound

This paper cites Discovering the Gems in Early Layers: Accelerating Long-Context LLMs with 1000x Input Token Reduction.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation Discovering the Gems in Early Layers: Accelerating Long-Context LLMs with 1000x Input Token Reduction

Reference 42

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.772628Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1bc901b6-2acf-4280-a233-9b90ced92faf · outbound

This paper cites Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism

Reference 43

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.776294Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 858ec0b8-1b25-42de-a93e-597ba1aa812a · outbound

This paper cites ShadowKV: KV Cache in Shadows for High-Throughput Long-Context LLM Inference.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation ShadowKV: KV Cache in Shadows for High-Throughput Long-Context LLM Inference

Reference 44

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.779908Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fbe6a1a6-a70f-4a57-83b8-747c165b2d18 · outbound

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

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation Gemini: A Family of Highly Capable Multimodal Models

Reference 45

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.783386Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 73bf3f2f-7191-4dfb-b041-4b4823b62010 · outbound

This paper cites LOOK-M: Look-Once Optimization in KV Cache for Efficient Multimodal Long-Context Inference.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation LOOK-M: Look-Once Optimization in KV Cache for Efficient Multimodal Long-Context Inference

Reference 46

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.788093Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e4b23d10-0562-4206-9311-5db5d06a788b · outbound

This paper cites Emergent Abilities of Large Language Models.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation Emergent Abilities of Large Language Models

Reference 47

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.791959Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 771287dc-540e-4263-a3ca-95401869bafc · outbound

This paper cites Unlocking Efficiency in Large Language Model Inference: A Comprehensive Survey of Speculative Decoding.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation Unlocking Efficiency in Large Language Model Inference: A Comprehensive Survey of Speculative Decoding

Reference 48

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.795746Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3fbfa1cb-6a3e-4d12-81c3-d6c13e579e7d · outbound

This paper cites DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads

Reference 49

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.799746Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ab3ff739-7f95-4d28-933e-196791966058 · outbound

This paper cites Efficient Streaming Language Models with Attention Sinks.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation Efficient Streaming Language Models with Attention Sinks

Reference 50

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.803412Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f80f3215-5206-4997-ac88-d9ea0cd44b21 · outbound

This paper cites Effective Long-Context Scaling of Foundation Models.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation Effective Long-Context Scaling of Foundation Models

Reference 51

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.807536Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 98b2bab2-103c-456d-90cf-932a6058636a · outbound

This paper cites Qwen2 Technical Report.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation Qwen2 Technical Report

Reference 52

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.811606Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ecfd4f90-ce9d-497b-a883-24e8b28f6620 · outbound

This paper cites LLM Inference Unveiled: Survey and Roofline Model Insights.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation LLM Inference Unveiled: Survey and Roofline Model Insights

Reference 53

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.815570Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fec84020-9065-4149-b785-b80c0267ae27 · outbound

This paper cites Draft & Verify: Lossless Large Language Model Acceleration via Self-Speculative Decoding.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation Draft & Verify: Lossless Large Language Model Acceleration via Self-Speculative Decoding

Reference 54

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.819360Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e183dc25-5ec7-4f8a-b359-10b893767903 · outbound

This paper cites H$_2$O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation H$_2$O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models

Reference 55

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.823352Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4df35ab7-1b69-492f-9ad8-f440963f86a7 · outbound

This paper cites Q-hitter: A better token oracle for efficient llm inference via sparse-quantized kv cache.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation Q-hitter: A better token oracle for efficient llm inference via sparse-quantized kv cache

Reference 56

Resolution: verified fuzzy; raw_fallback, observed 2026-08-09T11:11:30.978098Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 3cc57133-f199-4d1a-ab98-0d5a9cc4a8f4 · outbound

This paper cites SGLang: Efficient Execution of Structured Language Model Programs.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation SGLang: Efficient Execution of Structured Language Model Programs

Reference 57

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.832003Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9f5f087b-b25b-45c8-9ae1-ec36d0fe213d · outbound

This paper cites Instruction-Following Evaluation for Large Language Models.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation Instruction-Following Evaluation for Large Language Models

Reference 58

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.835920Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e00000cd-e7e5-4938-b067-d560714c60f6 · outbound

This paper cites SampleAttention: Near-Lossless Acceleration of Long Context LLM Inference with Adaptive Structured Sparse Attention.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation SampleAttention: Near-Lossless Acceleration of Long Context LLM Inference with Adaptive Structured Sparse Attention

Reference 59

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.839813Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation 68f5277a-7d78-4afb-813e-57c80128c8d0 · inbound

SALE : Low-bit Estimation for Efficient Sparse Attention in Long-context LLM Prefilling cites this paper.

SALE : Low-bit Estimation for Efficient Sparse Attention in Long-context LLM Prefilling Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-07T12:39:05.263720Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 559406c1-ef58-46ce-b620-3a64915e8109 · inbound

Multi-Modal Agents for Power Distribution Defect Detection: An Evaluation of Foundation Models cites this paper.

Multi-Modal Agents for Power Distribution Defect Detection: An Evaluation of Foundation Models Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation

Reference 47

Resolution: verified exact; arxiv_id, observed 2026-07-03T14:48:33.366365Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation bdf29ee7-e41d-4710-93d0-0ab622e72a68 · inbound

Non-vacuous Generalization Bounds for Reinforcement Learning with Verifiable Rewards cites this paper.

Non-vacuous Generalization Bounds for Reinforcement Learning with Verifiable Rewards Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-02T02:01:48.669846Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.