Paper Citation Record · LEDGER

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning

As of 18 August 2026, this Paper Citation Record lists 32 of 32 outbound references and 0 inbound Pith citation observations for arXiv:2506.22919.

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

pith.paper-citation-record.v1
2506.22919 v2

Coverage vector

measured 32 of 32 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-06T21:59:52.247745Z

measured 32 of 32 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 0 of 0 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links

measured 0 of 1 external citation measurements

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

Source: cited_works

Reference resolution

32 of 32 outbound references displayed

verified exact5
verified fuzzy10
unresolved16
parse uncertain0
malformed identifier1
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 60bdd10f-92ae-471f-a03a-a76dbbab877d · outbound

This paper cites Le, and Geoffrey Hinton.

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning Le, and Geoffrey Hinton

Reference 1

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T21:59:55.630496Z

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 309b2c77-a360-4830-851b-86e9fbb54a21 · outbound

This paper cites Switch transformers: Scaling to trillion parameter models with simple and efficient sparsity.Journal of Machine Learning Research, 23(120):1–39, 2022.

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning Switch transformers: Scaling to trillion parameter models with simple and efficient sparsity.Journal of Machine Learning Research, 23(120):1–39, 2022

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-06T21:59:49.606183Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0d8f1028-5cec-417d-bc00-149ebfb90d3e · outbound

This paper cites GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding.

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-06T21:59:49.709718Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d8dbf301-52b4-4a64-825b-30a86a5fd31d · outbound

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

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-06T21:59:49.832400Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1b363b70-7006-47b6-831d-004f74b004b4 · outbound

This paper cites Spatial Mixture-of-Experts.

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning Spatial Mixture-of-Experts

Reference 5

Resolution: verified exact; local_arxiv, observed 2026-08-06T21:59:53.127458Z

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 be5941ef-23e5-4c87-a329-368af8cb73fe · outbound

This paper cites Mixture-of-Experts with Expert Choice Routing.

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning Mixture-of-Experts with Expert Choice Routing

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-06T21:59:50.065384Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e256eb57-97a2-4c87-8379-42a0163b7606 · outbound

This paper cites Pre-gated MoE: An Algorithm-System Co-Design for Fast and Scalable Mixture-of-Expert Inference.

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning Pre-gated MoE: An Algorithm-System Co-Design for Fast and Scalable Mixture-of-Expert Inference

Reference 7

Resolution: verified exact; local_arxiv, observed 2026-08-06T21:59:52.966451Z

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 ba70103b-0b5a-4fe0-bf73-ccdcf82fb49b · outbound

This paper cites Spectra: Specialized experts from prompt tuning.

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning Spectra: Specialized experts from prompt tuning

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-06T21:59:50.227752Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3469579c-4014-4539-b71a-7695b16064fb · outbound

This paper cites MoFE: Mixture of Frozen Experts Architecture.

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning MoFE: Mixture of Frozen Experts Architecture

Reference 9

Resolution: verified exact; local_arxiv, observed 2026-08-06T21:59:52.757826Z

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 c59ac4f5-1fb6-458c-b1c8-f48982e6a70f · outbound

This paper cites ModuleFormer: Modularity Emerges from Mixture-of-Experts.

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning ModuleFormer: Modularity Emerges from Mixture-of-Experts

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-06T21:59:50.455762Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a7f9b936-c339-4e02-8aaf-9e9055d6dcb1 · outbound

This paper cites AutoMoE: Heterogeneous Mixture-of-Experts with Adaptive Computation for Efficient Neural Machine Translation.

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning AutoMoE: Heterogeneous Mixture-of-Experts with Adaptive Computation for Efficient Neural Machine Translation

Reference 11

Resolution: verified exact; local_arxiv, observed 2026-08-06T21:59:52.602236Z

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 cc7bfa8b-6dc4-47de-9f03-c8c844c7093f · outbound

This paper cites HetuMoE: An Efficient Trillion-scale Mixture-of-Expert Distributed Training System.

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning HetuMoE: An Efficient Trillion-scale Mixture-of-Expert Distributed Training System

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-06T21:59:50.741606Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a55f8727-180e-46d9-b188-4c0d8ee21fee · outbound

This paper cites Optimizing Distributed Deployment of Mixture-of-Experts Model Inference in Serverless Computing.

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning Optimizing Distributed Deployment of Mixture-of-Experts Model Inference in Serverless Computing

Reference 13

Resolution: verified exact; local_arxiv, observed 2026-08-06T21:59:52.473973Z

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 ffc2873b-1fa5-4fc2-b41e-70f1660cc949 · outbound

This paper cites Lynx: Enabling Efficient MoE Inference through Dynamic Batch-Aware Expert Selection.

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning Lynx: Enabling Efficient MoE Inference through Dynamic Batch-Aware Expert Selection

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-06T21:59:50.990181Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2daa65f7-25ab-441c-8a07-84edde042d01 · outbound

This paper cites Mixture of Experts Made Intrinsically Interpretable.

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning Mixture of Experts Made Intrinsically Interpretable

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-06T21:59:51.114219Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e60790ab-ee19-4e09-a1ef-42a9d02db857 · outbound

This paper cites DeepSpeed-MoE: Advancing Mixture-of-Experts Inference and Training to Power Next-Generation AI Scale.

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning DeepSpeed-MoE: Advancing Mixture-of-Experts Inference and Training to Power Next-Generation AI Scale

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-06T21:59:51.170050Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 044cc1e5-d93a-40d2-ac74-686209d5134c · outbound

This paper cites ExpertRAG: Efficient RAG with Mixture of Experts -- Optimizing Context Retrieval for Adaptive LLM Responses.

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning ExpertRAG: Efficient RAG with Mixture of Experts -- Optimizing Context Retrieval for Adaptive LLM Responses

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-06T21:59:51.219171Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0b6b33a2-3f54-4fbd-9438-dc22301a1ed2 · outbound

This paper cites DeRS: Towards Extremely Efficient Upcycled Mixture-of-Experts Models.

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning DeRS: Towards Extremely Efficient Upcycled Mixture-of-Experts Models

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-06T21:59:51.254870Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 62c484b2-9f77-49b7-83df-7e7153903605 · outbound

This paper cites Manning, Andrew Ng, and Christopher Potts.

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning Manning, Andrew Ng, and Christopher Potts

Reference 19

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T21:59:55.249289Z

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 41bfd936-739a-4324-88b6-1a8545c93b0e · outbound

This paper cites Gating is Weighting: Understanding Gated Linear Attention through In-context Learning.

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning Gating is Weighting: Understanding Gated Linear Attention through In-context Learning

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-06T21:59:51.366299Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bba7f7c2-dd04-4b92-8ba4-29e4791a7706 · outbound

This paper cites MoESys: A Distributed and Efficient Mixture-of-Experts Training and Inference System for Internet Services.

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning MoESys: A Distributed and Efficient Mixture-of-Experts Training and Inference System for Internet Services

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-06T21:59:51.427121Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 467ffaca-465b-4b9d-b1a1-7dc41a937277 · outbound

This paper cites MoLEx: Mixture of Layer Experts for Finetuning with Sparse Upcycling.

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning MoLEx: Mixture of Layer Experts for Finetuning with Sparse Upcycling

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-06T21:59:51.527234Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation be2399fe-57d2-4b30-ab5a-3fab6c346652 · outbound

This paper cites Two-expert routing adds a modest +0.58pp over Top–1, still below the GRU+GRU upper- bound (90.62%).

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning Two-expert routing adds a modest +0.58pp over Top–1, still below the GRU+GRU upper- bound (90.62%)

Reference 23

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T21:59:55.019473Z

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 cf6e564b-14f0-49cb-957e-9aaf43c8ee4b · outbound

This paper cites Entropy plummets from 0.65 to 0.15 bits, indicating almost deterministic gating.

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning Entropy plummets from 0.65 to 0.15 bits, indicating almost deterministic gating

Reference 24

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T21:59:54.651214Z

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 1ca250bc-f7a4-4ddf-bdd2-16dbebc8a751 · outbound

This paper cites always run the GRU.

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning always run the GRU

Reference 25

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T21:59:54.272552Z

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 7ea3c8ce-77c0-445d-bd01-b461f07fa6db · outbound

This paper cites Because two experts fire per sample, single-expert attribution is lost - diluting the inter- pretability benefit of sparse MoEs.

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning Because two experts fire per sample, single-expert attribution is lost - diluting the inter- pretability benefit of sparse MoEs

Reference 26

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T21:59:53.922369Z

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 bdb4b1ab-253d-43b1-ad49-de25eab26a27 · outbound

This paper cites Executing both lightweight heads adds < 0.1 ms on a T4 GPU; compute cost is negligible.

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning Executing both lightweight heads adds < 0.1 ms on a T4 GPU; compute cost is negligible

Reference 27

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T21:59:53.882957Z

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 ba82faa4-c5a1-4229-9187-39e6a55a3b6b · outbound

This paper cites Removing both losses scarcely changes accuracy (+0.13pp difference).

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning Removing both losses scarcely changes accuracy (+0.13pp difference)

Reference 28

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T21:59:53.761905Z

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 1879f162-15cd-469b-9782-76d4f4cce305 · outbound

This paper cites Entropy collapses from 0.40 bits to 0.02 bits—the gate becomes almost deterministic.

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning Entropy collapses from 0.40 bits to 0.02 bits—the gate becomes almost deterministic

Reference 29

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T21:59:53.627356Z

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 efefa1ed-897c-4a86-8353-bf476bb6d453 · outbound

This paper cites Without losses, the FFNN path receives 99.7% of inputs; the GRU is effectively bypassed.

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning Without losses, the FFNN path receives 99.7% of inputs; the GRU is effectively bypassed

Reference 30

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T21:59:53.514368Z

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 f1e995ce-7be8-48a0-a5dd-591956d02d6a · outbound

This paper cites Regularisation preserves a bal- anced and interpretable routing pattern, while the un-regularised variant hides all computation in one expert.

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning Regularisation preserves a bal- anced and interpretable routing pattern, while the un-regularised variant hides all computation in one expert

Reference 31

Resolution: malformed identifier; raw_fallback, observed 2026-08-06T21:59:53.398974Z

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 4415ca7f-80a4-49bf-a6e6-f7ea3a0cccc8 · outbound

This paper cites an unresolved cited work.

Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning Unresolved cited work

Reference 32

Resolution: unresolved; raw_fallback, observed 2026-08-06T21:59:53.261853Z

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.

Pith citing papers

No inbound Pith citation observations are available.