Paper Citation Record · LEDGER

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity

As of 18 August 2026, this Paper Citation Record lists 67 of 67 outbound references and 2 inbound Pith citation observations for arXiv:2507.08771.

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

pith.paper-citation-record.v1
2507.08771 v2

Coverage vector

measured 67 of 67 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-06T18:20:00.523197Z

measured 69 of 69 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-18T06:34:40.430872+00:00

measured 2 of 2 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-06-28T22:50:51.900169Z

measured 0 of 1 external citation measurements

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

Source: arxiv_reference, observed 2026-07-04T15:39:56.501546Z

Reference resolution

67 of 67 outbound references displayed

verified exact0
verified fuzzy24
unresolved43
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 25fa0fb7-6e23-4bfd-8cc3-f9f36670b563 · outbound

This paper cites PIQA : Reasoning about physical commonsense in natural language.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity PIQA : Reasoning about physical commonsense in natural language

Reference 1

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:20:02.033486Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation b57dedeb-ae76-41c5-b3a2-f55d69df0f44 · outbound

This paper cites Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.123050Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d59b9e6c-aee0-49b5-b253-a7558066ea20 · outbound

This paper cites BoolQ : Exploring the surprising difficulty of natural yes/no questions.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity BoolQ : Exploring the surprising difficulty of natural yes/no questions

Reference 3

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:20:02.006520Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 2d36ce6a-fa90-48ad-8658-51e69906c554 · outbound

This paper cites TyDi QA : A benchmark for information-seeking question answering in typologically diverse languages.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity TyDi QA : A benchmark for information-seeking question answering in typologically diverse languages

Reference 4

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:20:01.964734Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation fc253894-037c-4f0e-9518-3778fdd8336b · outbound

This paper cites DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.139543Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b66a0274-5a9c-40fd-a5c4-936255cadfbd · outbound

This paper cites Language modeling with gated convolutional networks.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity Language modeling with gated convolutional networks

Reference 6

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:20:01.926715Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation becf4859-0d6b-47cb-a1c7-33fad50d7330 · outbound

This paper cites Enhancing Chat Language Models by Scaling High-quality Instructional Conversations.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity Enhancing Chat Language Models by Scaling High-quality Instructional Conversations

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.152649Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 086fbc5d-070c-4a7f-97f1-51f4fe6b4ca0 · outbound

This paper cites The Llama 3 Herd of Models.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity The Llama 3 Herd of Models

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.158485Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 20c700f7-2af5-4d32-9a65-1da0ef4ef76f · outbound

This paper cites Switch Transformers : Scaling to trillion parameter models with simple and efficient sparsity.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity Switch Transformers : Scaling to trillion parameter models with simple and efficient sparsity

Reference 9

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:20:01.905046Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 8dd14c3c-02c3-43d8-a52c-63e68d5035e8 · outbound

This paper cites SparseGPT : Massive language models can be accurately pruned in one-shot.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity SparseGPT : Massive language models can be accurately pruned in one-shot

Reference 10

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:20:01.877043Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 9883f86a-5a9f-4832-bfc4-db0bda45f159 · outbound

This paper cites MegaBlocks : Efficient sparse training with mixture-of-experts.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity MegaBlocks : Efficient sparse training with mixture-of-experts

Reference 11

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:20:01.851894Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 709b6dab-f0d3-4921-a1af-d33a2f5ba256 · outbound

This paper cites The Pile: An 800GB Dataset of Diverse Text for Language Modeling.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity The Pile: An 800GB Dataset of Diverse Text for Language Modeling

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.176813Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e61f35ce-cdea-4b3b-83f8-1e1b304a7f98 · outbound

This paper cites MiniLLM: On-Policy Distillation of Large Language Models.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity MiniLLM: On-Policy Distillation of Large Language Models

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.185451Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8f82d564-9f3c-467a-876f-ff097377227b · outbound

This paper cites FastMoE: A Fast Mixture-of-Expert Training System.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity FastMoE: A Fast Mixture-of-Expert Training System

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.189812Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4885e582-1654-482f-ba39-538194b9bca2 · outbound

This paper cites Distilling Step-by-Step! Outperforming Larger Language Models with Less Training Data and Smaller Model Sizes.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity Distilling Step-by-Step! Outperforming Larger Language Models with Less Training Data and Smaller Model Sizes

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.195091Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c560bb00-3ef7-4756-a726-d4c3e909313b · outbound

This paper cites MiniCPM: Unveiling the Potential of Small Language Models with Scalable Training Strategies.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity MiniCPM: Unveiling the Potential of Small Language Models with Scalable Training Strategies

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.199417Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e4a7d469-e966-483b-81f5-98417631cdf9 · outbound

This paper cites Harder tasks need more experts: Dynamic routing in MoE models.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity Harder tasks need more experts: Dynamic routing in MoE models

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.204257Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 58fe1ec6-c0e1-40e4-9fae-4a7ee2be6c00 · outbound

This paper cites Tutel: Adaptive mixture-of-experts at scale.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity Tutel: Adaptive mixture-of-experts at scale

Reference 18

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:20:01.834840Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 1b3dc5d7-1be0-492d-8bb4-a90ea8d92b04 · outbound

This paper cites Mixtral of Experts.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity Mixtral of Experts

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.213261Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7060e794-53d6-4614-9403-522dba46b907 · outbound

This paper cites Scaling Laws for Fine-Grained Mixture of Experts.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity Scaling Laws for Fine-Grained Mixture of Experts

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.225860Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e2e6cce0-03e7-4d88-9e45-03f35124285f · outbound

This paper cites Fast inference from Transformers via speculative decoding.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity Fast inference from Transformers via speculative decoding

Reference 21

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:20:01.818992Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation a0488d49-b902-4415-a79d-6c3bb7b0083d · outbound

This paper cites StarCoder: may the source be with you!.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity StarCoder: may the source be with you!

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.239663Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 516aafb3-7494-4d74-b9cd-4b1998290d42 · outbound

This paper cites EAGLE: Speculative Sampling Requires Rethinking Feature Uncertainty.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity EAGLE: Speculative Sampling Requires Rethinking Feature Uncertainty

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.246316Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c78072f2-a450-4ca3-9f8a-8033e5a7460d · outbound

This paper cites EAGLE-2 : Faster inference of language models with dynamic draft trees.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity EAGLE-2 : Faster inference of language models with dynamic draft trees

Reference 24

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:20:01.800258Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation d03b74b2-0cf4-4847-bd92-31c8cb1afab9 · outbound

This paper cites The lazy neuron phenomenon: On emergence of activation sparsity in Transformers.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity The lazy neuron phenomenon: On emergence of activation sparsity in Transformers

Reference 25

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:20:01.785310Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation f4f91eca-3bb6-465d-aeff-540963d4cddc · outbound

This paper cites DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.291096Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e398f203-85dc-44a6-81fa-8cc2a9e7ff9a · outbound

This paper cites DeepSeek-V3 Technical Report.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity DeepSeek-V3 Technical Report

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.296637Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 47be00ea-9acd-4eb0-a587-56d76c9d458e · outbound

This paper cites GRIN: GRadient-INformed MoE.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity GRIN: GRadient-INformed MoE

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.301919Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e7468f16-caaa-4d13-a243-fb27d8bae376 · outbound

This paper cites Deja Vu : Contextual sparsity for efficient LLMs at inference time.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity Deja Vu : Contextual sparsity for efficient LLMs at inference time

Reference 29

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:20:01.771121Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 72eb8b80-925b-4b3b-8508-d59bd534c81f · outbound

This paper cites Sparsing Law: Towards Large Language Models with Greater Activation Sparsity.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity Sparsing Law: Towards Large Language Models with Greater Activation Sparsity

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.312851Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2e3d2b29-e9fa-4350-8b6a-64625e583a2e · outbound

This paper cites LLM-Pruner: On the Structural Pruning of Large Language Models.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity LLM-Pruner: On the Structural Pruning of Large Language Models

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.319815Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d7dd08c2-2153-43ea-a3d4-cf8960f2ad0a · outbound

This paper cites ReLU Strikes Back: Exploiting Activation Sparsity in Large Language Models.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity ReLU Strikes Back: Exploiting Activation Sparsity in Large Language Models

Reference 32

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.326711Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1c792a5f-a297-4f88-a602-aff13ae896e4 · outbound

This paper cites Soft Merging of Experts with Adaptive Routing.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity Soft Merging of Experts with Adaptive Routing

Reference 33

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.332298Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 849baa19-38b4-4075-bc2a-937cad948e88 · outbound

This paper cites The LAMBADA dataset: Word prediction requiring a broad discourse context.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity The LAMBADA dataset: Word prediction requiring a broad discourse context

Reference 34

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:20:01.755957Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 575f85a8-5179-4854-a603-839825c544cb · outbound

This paper cites Exploring the limits of transfer learning with a unified text-to-text Transformer.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity Exploring the limits of transfer learning with a unified text-to-text Transformer

Reference 35

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:20:01.739615Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation fd5e66af-7322-43de-a68a-59eb8cc327b5 · outbound

This paper cites Improving Dictionary Learning with Gated Sparse Autoencoders.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity Improving Dictionary Learning with Gated Sparse Autoencoders

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.348641Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7dfb9ccb-c0d2-412c-955c-cf9a429fa4c8 · outbound

This paper cites Searching for Activation Functions.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity Searching for Activation Functions

Reference 37

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.353233Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8ad04899-f78c-420b-afd3-cc161c9d8e16 · outbound

This paper cites SocialIQA : Commonsense reasoning about social interactions.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity SocialIQA : Commonsense reasoning about social interactions

Reference 38

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:20:01.724586Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 9d3363eb-a434-4907-826e-0a48c4902190 · outbound

This paper cites OmniQuant: Omnidirectionally Calibrated Quantization for Large Language Models.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity OmniQuant: Omnidirectionally Calibrated Quantization for Large Language Models

Reference 39

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.363844Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a608e87e-f83b-4a60-8507-4b651100694d · outbound

This paper cites GLU Variants Improve Transformer.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity GLU Variants Improve Transformer

Reference 40

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.369351Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 20f5ddc7-99d1-45d0-8b5b-cd3b6b779129 · outbound

This paper cites Dolma: an Open Corpus of Three Trillion Tokens for Language Model Pretraining Research.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity Dolma: an Open Corpus of Three Trillion Tokens for Language Model Pretraining Research

Reference 41

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.374027Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b190fc45-d9e5-4ddf-9ad9-dac4c51e70e5 · outbound

This paper cites P ro S parse: Introducing and enhancing intrinsic activation sparsity within large language models.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity P ro S parse: Introducing and enhancing intrinsic activation sparsity within large language models

Reference 42

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:20:01.697972Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 0aeb7da2-92c1-4501-bf32-4b3b6ad58c68 · outbound

This paper cites PowerInfer: Fast Large Language Model Serving with a Consumer-grade GPU.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity PowerInfer: Fast Large Language Model Serving with a Consumer-grade GPU

Reference 43

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.384466Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e03d1162-e296-44da-a76d-51bfbd2de612 · outbound

This paper cites Turbo Sparse: Achieving LLM SOTA Performance with Minimal Activated Parameters.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity Turbo Sparse: Achieving LLM SOTA Performance with Minimal Activated Parameters

Reference 44

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.389662Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2649523e-6833-4eb3-ae6b-9b1de0c4042f · outbound

This paper cites A Simple and Effective Pruning Approach for Large Language Models.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity A Simple and Effective Pruning Approach for Large Language Models

Reference 45

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.395583Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 365695a0-91c8-4964-a070-4aa6c8fbb17a · outbound

This paper cites CUTLASS , Jan 2023.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity CUTLASS , Jan 2023

Reference 46

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:20:01.682402Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation aaabc245-9903-425a-a86a-2377bfe0ce71 · outbound

This paper cites Efficient large language models: A survey.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity Efficient large language models: A survey

Reference 47

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:20:01.661552Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 5e66add9-f76c-47ba-ab8f-76dc9a53554e · outbound

This paper cites Auxiliary-Loss-Free Load Balancing Strategy for Mixture-of-Experts.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity Auxiliary-Loss-Free Load Balancing Strategy for Mixture-of-Experts

Reference 48

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.418545Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ad462045-dcd7-40d1-8b27-cfd91c2ebd0c · outbound

This paper cites ReMoE: Fully Differentiable Mixture-of-Experts with ReLU Routing.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity ReMoE: Fully Differentiable Mixture-of-Experts with ReLU Routing

Reference 49

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.424283Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cece94a5-82ba-4719-b580-25b15f3b141d · outbound

This paper cites Magicoder: Empowering Code Generation with OSS-Instruct.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity Magicoder: Empowering Code Generation with OSS-Instruct

Reference 50

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.429721Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4934800a-3021-4a03-83c0-93cf3af0f586 · outbound

This paper cites Unlocking efficiency in large language model inference: A comprehensive survey of speculative decoding.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity Unlocking efficiency in large language model inference: A comprehensive survey of speculative decoding

Reference 51

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:20:01.644295Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 3a1fcbb4-978f-4882-940b-574372928e36 · outbound

This paper cites Sheared LLaMA: Accelerating Language Model Pre-training via Structured Pruning.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity Sheared LLaMA: Accelerating Language Model Pre-training via Structured Pruning

Reference 52

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.442308Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e4fe0660-d3de-4f05-8bcb-269058477b56 · outbound

This paper cites Smoothquant: Accurate and efficient post-training quantization for large language models.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity Smoothquant: Accurate and efficient post-training quantization for large language models

Reference 53

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:20:01.627328Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 9b26c98d-7cb2-413b-9054-b3e19303c0fa · outbound

This paper cites WizardLM: Empowering large pre-trained language models to follow complex instructions.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity WizardLM: Empowering large pre-trained language models to follow complex instructions

Reference 54

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.452578Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c3bd469b-6ec3-40fb-a33e-85f97cc595f6 · outbound

This paper cites PowerInfer-2: Fast Large Language Model Inference on a Smartphone.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity PowerInfer-2: Fast Large Language Model Inference on a Smartphone

Reference 55

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.457313Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a97201bc-28ae-4c43-ab1a-f42024abd097 · outbound

This paper cites Tensor Programs V: Tuning Large Neural Networks via Zero-Shot Hyperparameter Transfer.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity Tensor Programs V: Tuning Large Neural Networks via Zero-Shot Hyperparameter Transfer

Reference 56

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.461750Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 25797477-e26a-421e-9ef0-e44c03e8bedc · outbound

This paper cites ZeroQuant-V2: Exploring Post-training Quantization in LLMs from Comprehensive Study to Low Rank Compensation.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity ZeroQuant-V2: Exploring Post-training Quantization in LLMs from Comprehensive Study to Low Rank Compensation

Reference 57

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.467026Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a277af18-f8a9-4d11-942e-f0bcdd3958cc · outbound

This paper cites HellaSwag : Can a machine really finish your sentence? In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pp.\ 4791--4800, 2019.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity HellaSwag : Can a machine really finish your sentence? In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pp.\ 4791--4800, 2019

Reference 58

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:20:01.607219Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation c5e6c3cf-3197-4da6-b48a-e310b78750f1 · outbound

This paper cites Root mean square layer normalization.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity Root mean square layer normalization

Reference 59

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:20:01.586515Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 01cbe2a3-37ea-4fb4-8e35-e1dc8e9e3ef6 · outbound

This paper cites ReLU$^2$ Wins: Discovering Efficient Activation Functions for Sparse LLMs.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity ReLU$^2$ Wins: Discovering Efficient Activation Functions for Sparse LLMs

Reference 60

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.484184Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c9db0349-ea9d-4d1e-b7d4-e71de08952b2 · outbound

This paper cites Exploring the benefit of activation sparsity in pre-training.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity Exploring the benefit of activation sparsity in pre-training

Reference 61

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:20:01.569359Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 516e2264-3414-4840-8cbc-e056a509ae59 · outbound

This paper cites Ouroboros: Generating longer drafts phrase by phrase for faster speculative decoding.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity Ouroboros: Generating longer drafts phrase by phrase for faster speculative decoding

Reference 62

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T18:20:01.552836Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 3d8819fc-ab7a-49f8-aae5-4dc2a5f00633 · outbound

This paper cites FR-Spec: Accelerating Large-Vocabulary Language Models via Frequency-Ranked Speculative Sampling.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity FR-Spec: Accelerating Large-Vocabulary Language Models via Frequency-Ranked Speculative Sampling

Reference 63

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.499895Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ec30a20a-b760-4d15-8bf7-5a8616bae220 · outbound

This paper cites Lory: Fully Differentiable Mixture-of-Experts for Autoregressive Language Model Pre-training.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity Lory: Fully Differentiable Mixture-of-Experts for Autoregressive Language Model Pre-training

Reference 64

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.505281Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 11de79ca-38ea-4612-a47d-3447bd797cfc · outbound

This paper cites @esa (Ref.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity @esa (Ref

Reference 65

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.510593Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1a79635e-532f-4ff6-88bd-ab19558b5e33 · outbound

This paper cites an unresolved cited work.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity Unresolved cited work

Reference 66

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.517604Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cefe486b-021e-4a95-a1ab-6965acf4db97 · outbound

This paper cites an unresolved cited work.

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity Unresolved cited work

Reference 67

Resolution: unresolved; no resolver link, observed 2026-08-06T18:20:00.523197Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation 443bc417-9566-4821-9188-faa0047dcff7 · inbound

dMoE: dLLMs with Learnable Block Experts cites this paper.

dMoE: dLLMs with Learnable Block Experts BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity

Reference 21

Resolution: verified exact; arxiv_id, observed 2026-06-28T22:52:45.095922Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 5527f436-88e0-4a7d-bd8e-e1decd92f1cc · inbound

GeMoE: Gating Entropy is All You Need for Uncertainty-aware Adaptive Routing in MoE-based Large Vision-Language Models cites this paper.

GeMoE: Gating Entropy is All You Need for Uncertainty-aware Adaptive Routing in MoE-based Large Vision-Language Models BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity

Reference 51

Resolution: verified exact; arxiv_id, observed 2026-07-04T15:39:56.503036Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.