Paper Citation Record · LEDGER

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation

As of 7 August 2026, this Paper Citation Record lists 42 of 42 outbound references and 0 inbound Pith citation observations for arXiv:2606.17518.

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

pith.paper-citation-record.v1
2606.17518 v1

Coverage vector

measured 42 of 42 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-06-26T23:35:32.175768Z

measured 42 of 42 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-07T06:34:17.273281+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

42 of 42 outbound references displayed

verified exact23
verified fuzzy0
unresolved16
parse uncertain0
malformed identifier0
metadata mismatch3

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 6989585d-35a4-4d6d-814e-6319a389340b · outbound

This paper cites Hydra: Sequentially-Dependent Draft Heads for Medusa Decoding.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation Hydra: Sequentially-Dependent Draft Heads for Medusa Decoding

Reference 1

Resolution: verified exact; arxiv_id, observed 2026-07-03T22:29:00.460659Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 0043db8c-d393-48c1-9038-9a0667192661 · outbound

This paper cites Kevin: Multi-Turn RL for Generating CUDA Kernels.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation Kevin: Multi-Turn RL for Generating CUDA Kernels

Reference 2

Resolution: verified exact; arxiv_id, observed 2026-07-03T22:29:00.389666Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 85fb37fd-f540-4bac-8e0c-247538a7e88a · outbound

This paper cites Dynamic Depth Decoding: Faster Speculative Decoding for LLMs.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation Dynamic Depth Decoding: Faster Speculative Decoding for LLMs

Reference 3

Resolution: verified exact; arxiv_id, observed 2026-07-03T22:29:00.438374Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 7b29fc6e-9ba8-4e09-8d32-18460673c192 · outbound

This paper cites Lee, Deming Chen, and Tri Dao.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation Lee, Deming Chen, and Tri Dao

Reference 4

Resolution: unresolved; no resolver link, observed 2026-06-26T23:35:32.175768Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 54ec0ffb-4b83-4699-ba37-da27818ca63f · outbound

This paper cites K-search: Llm kernel generation via co-evolving intrinsic world model.arXiv preprint arXiv:2602.19128.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation K-search: Llm kernel generation via co-evolving intrinsic world model.arXiv preprint arXiv:2602.19128

Reference 5

Resolution: metadata mismatch; arxiv_id, observed 2026-07-03T22:29:00.451374Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation ba349331-901a-41f9-b383-8cad9ef64fd2 · outbound

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

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation Accelerating Large Language Model Decoding with Speculative Sampling

Reference 6

Resolution: verified exact; local_arxiv, observed 2026-07-03T22:29:00.400188Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 10980dc3-95f0-4422-90f9-cf80e15bf554 · outbound

This paper cites Cuda agent: Large-scale agentic rl for high-performance cuda kernel generation.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation Cuda agent: Large-scale agentic rl for high-performance cuda kernel generation

Reference 7

Resolution: verified exact; arxiv_id, observed 2026-07-03T22:29:00.431421Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 617c7d86-c4d4-4f8f-b679-405d69f602c3 · outbound

This paper cites Flashattention-2: Faster attention with better parallelism and work partitioning.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation Flashattention-2: Faster attention with better parallelism and work partitioning

Reference 8

Resolution: unresolved; no resolver link, observed 2026-06-26T23:35:32.175768Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b0ccec67-9360-45f4-9ea8-2a988f4058c1 · outbound

This paper cites Fu, Stefano Ermon, Atri Rudra, and Christopher Ré.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation Fu, Stefano Ermon, Atri Rudra, and Christopher Ré

Reference 9

Resolution: unresolved; no resolver link, observed 2026-06-26T23:35:32.175768Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation dfa07c84-9c65-49df-8a87-e0221fc63ab2 · outbound

This paper cites DeepSeek-V4: Towards highly efficient million-token context intelligence.https://huggingface.co/deepseek-ai/DeepSeek- V4-Pro, 2026.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation DeepSeek-V4: Towards highly efficient million-token context intelligence.https://huggingface.co/deepseek-ai/DeepSeek- V4-Pro, 2026

Reference 10

Resolution: unresolved; no resolver link, observed 2026-06-26T23:35:32.175768Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3fd192cd-30ea-487c-902c-6c8bfb3406ea · outbound

This paper cites STARK: Strategic team of agents for refining kernels.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation STARK: Strategic team of agents for refining kernels

Reference 11

Resolution: verified exact; arxiv_id, observed 2026-07-03T22:29:00.410602Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation bc1ad1f9-c2ee-406c-a180-ee319c28712e · outbound

This paper cites Ker- nelBlaster: Continual cross-task CUDA optimization via memory- augmented in-context reinforcement learning.CoRR, abs/2602.14293.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation Ker- nelBlaster: Continual cross-task CUDA optimization via memory- augmented in-context reinforcement learning.CoRR, abs/2602.14293

Reference 12

Resolution: verified exact; arxiv_id, observed 2026-07-03T22:29:00.445307Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 7b911a36-24df-48b9-84f7-e84a7aa66123 · outbound

This paper cites Kernel-Smith: A Unified Recipe for Evolutionary Kernel Optimization.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation Kernel-Smith: A Unified Recipe for Evolutionary Kernel Optimization

Reference 13

Resolution: verified exact; local_arxiv, observed 2026-07-03T22:29:00.435123Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 5bd2880f-54f5-40fc-830d-8ea85ce9661a · outbound

This paper cites GLM-5: from Vibe Coding to Agentic Engineering.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation GLM-5: from Vibe Coding to Agentic Engineering

Reference 14

Resolution: verified exact; local_arxiv, observed 2026-07-03T22:29:00.382781Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation bc7e9fa6-0c6f-4d26-9963-7a6819a7b9ab · outbound

This paper cites DRTriton: Large-Scale Synthetic Data Driven Reinforcement Learning for Triton Kernel Generation.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation DRTriton: Large-Scale Synthetic Data Driven Reinforcement Learning for Triton Kernel Generation

Reference 15

Resolution: verified exact; local_arxiv, observed 2026-07-03T22:29:00.442252Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation a9d1a71c-2a4a-4f9d-a785-3475cf7054cd · outbound

This paper cites arXiv preprint arXiv:2603.29010v1 , year =.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation arXiv preprint arXiv:2603.29010v1 , year =

Reference 16

Resolution: metadata mismatch; arxiv_id, observed 2026-07-03T22:29:00.437719Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 1e915c14-8984-41f0-a661-49d229f163c7 · outbound

This paper cites AutoKernel: Autonomous GPU kernel optimization via iterative agent-driven search.arXiv preprint arXiv:2603.21331, 2026.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation AutoKernel: Autonomous GPU kernel optimization via iterative agent-driven search.arXiv preprint arXiv:2603.21331, 2026

Reference 17

Resolution: verified exact; arxiv_id, observed 2026-07-03T22:29:00.415663Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 281ec68d-46a8-48ee-8543-e08063103df6 · outbound

This paper cites Efficient memory management for large language model serving with pagedattention.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation Efficient memory management for large language model serving with pagedattention

Reference 18

Resolution: unresolved; no resolver link, observed 2026-06-26T23:35:32.175768Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 95aa359f-2511-4250-86dd-7738ecf31c57 · outbound

This paper cites Fast inference from transformers via speculative decoding.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation Fast inference from transformers via speculative decoding

Reference 19

Resolution: unresolved; no resolver link, observed 2026-06-26T23:35:32.175768Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 47bed10c-00a4-4ed1-be62-c062518bffed · outbound

This paper cites Tritonforge: Profiling-guided framework for automated triton kernel optimization.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation Tritonforge: Profiling-guided framework for automated triton kernel optimization

Reference 20

Resolution: verified exact; arxiv_id, observed 2026-07-03T22:29:00.454565Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation bec37e3e-2ffa-4f45-ad74-b72b3b25a64d · outbound

This paper cites Tritonbench: Benchmarking large language model capabilities for generating triton operators.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation Tritonbench: Benchmarking large language model capabilities for generating triton operators

Reference 21

Resolution: unresolved; no resolver link, observed 2026-06-26T23:35:32.175768Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 49c67a07-7e43-414a-8557-31a1c6b03299 · outbound

This paper cites CUDA-L1: Improving CUDA Optimization via Contrastive Reinforcement Learning.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation CUDA-L1: Improving CUDA Optimization via Contrastive Reinforcement Learning

Reference 22

Resolution: verified exact; arxiv_id, observed 2026-07-14T02:20:17.802860Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 0f908925-2efb-4679-a99d-2e7837973829 · outbound

This paper cites EAGLE: speculative sampling requires rethinking feature uncertainty.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation EAGLE: speculative sampling requires rethinking feature uncertainty

Reference 23

Resolution: unresolved; no resolver link, observed 2026-06-26T23:35:32.175768Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 44f330b5-a54b-48b1-8b27-82963a9c4334 · outbound

This paper cites Kangaroo: Lossless self-speculative decod- ing for accelerating llms via double early exiting.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation Kangaroo: Lossless self-speculative decod- ing for accelerating llms via double early exiting

Reference 24

Resolution: unresolved; no resolver link, observed 2026-06-26T23:35:32.175768Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 272edc7c-1a53-4f6c-8011-26df811f4264 · outbound

This paper cites Under review.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation Under review

Reference 25

Resolution: verified exact; arxiv_id, observed 2026-07-03T22:29:00.427200Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 5b7e1aa2-11f1-42b5-8970-bb7908eafef3 · outbound

This paper cites KernelAgent: Autonomous GPU ker- nel generation & optimization via deep agents, 2025.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation KernelAgent: Autonomous GPU ker- nel generation & optimization via deep agents, 2025

Reference 26

Resolution: unresolved; no resolver link, observed 2026-06-26T23:35:32.175768Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8002a6fd-3c25-499a-be75-7aea23bc6bd0 · outbound

This paper cites AlphaEvolve: A coding agent for scientific and algorithmic discovery.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation AlphaEvolve: A coding agent for scientific and algorithmic discovery

Reference 27

Resolution: verified exact; local_arxiv, observed 2026-07-03T22:29:00.383445Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 2f6e46b5-5bfb-4779-ada8-389571e7437b · outbound

This paper cites Zhang, William Hu, Christopher Ré, and Azalia Mirhoseini.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation Zhang, William Hu, Christopher Ré, and Azalia Mirhoseini

Reference 28

Resolution: unresolved; no resolver link, observed 2026-06-26T23:35:32.175768Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ff33ba64-4abc-40c4-8694-ddcbddf98ac1 · outbound

This paper cites Mooncake: Trad- ing more storage for less computation - A kvcache-centric architecture for serving LLM chatbot.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation Mooncake: Trad- ing more storage for less computation - A kvcache-centric architecture for serving LLM chatbot

Reference 29

Resolution: unresolved; no resolver link, observed 2026-06-26T23:35:32.175768Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 43620738-9437-4e2d-95d8-b72f101ac045 · outbound

This paper cites CuTeGen: An LLM-Based Agentic Framework for Generation and Optimization of High-Performance GPU Kernels using CuTe.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation CuTeGen: An LLM-Based Agentic Framework for Generation and Optimization of High-Performance GPU Kernels using CuTe

Reference 30

Resolution: verified exact; local_arxiv, observed 2026-07-03T22:29:00.403611Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation e3574a69-1050-45ee-bfd0-fa21e7089e95 · outbound

This paper cites Flashattention-3: Fast and accurate attention with asynchrony and low-precision.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation Flashattention-3: Fast and accurate attention with asynchrony and low-precision

Reference 31

Resolution: unresolved; no resolver link, observed 2026-06-26T23:35:32.175768Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation aa3763cd-2a9d-456c-9e3c-a2cfedb9356a · outbound

This paper cites Openevolve: an open-source evolutionary coding agent, 2025.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation Openevolve: an open-source evolutionary coding agent, 2025

Reference 32

Resolution: unresolved; no resolver link, observed 2026-06-26T23:35:32.175768Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a34d9a09-e606-4a4a-a3e2-8ffa7643091c · outbound

This paper cites Kernelskill: A multi-agent framework for gpu kernel optimization.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation Kernelskill: A multi-agent framework for gpu kernel optimization

Reference 33

Resolution: verified exact; arxiv_id, observed 2026-07-03T22:29:00.421261Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 8efa09c6-9076-47b9-805b-659a7d033b6b · outbound

This paper cites KernelEvolve: Scaling Agentic Kernel Coding for Heterogeneous AI Accelerators at Meta.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation KernelEvolve: Scaling Agentic Kernel Coding for Heterogeneous AI Accelerators at Meta

Reference 34

Resolution: verified exact; arxiv_id, observed 2026-07-08T02:18:09.597449Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 0d049f41-677d-4626-a6c6-1862e415561e · outbound

This paper cites Fine-tuning gpt-5 for gpu kernel generation.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation Fine-tuning gpt-5 for gpu kernel generation

Reference 35

Resolution: verified exact; arxiv_id, observed 2026-07-03T22:29:00.434840Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 4a565637-34fa-4324-8f73-ba9102e76de5 · outbound

This paper cites SwizzlePerf: Hardware-Aware LLMs for GPU Kernel Performance Optimization.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation SwizzlePerf: Hardware-Aware LLMs for GPU Kernel Performance Optimization

Reference 36

Resolution: metadata mismatch; arxiv_id, observed 2026-07-03T22:29:00.457665Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation f6acc69a-26d1-4b5a-85e7-bbf95d9cade6 · outbound

This paper cites Astra: A multi-agent system for GPU kernel performance optimization.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation Astra: A multi-agent system for GPU kernel performance optimization

Reference 37

Resolution: verified exact; arxiv_id, observed 2026-07-03T22:29:00.409785Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation f22b2ca6-2561-4026-b5c1-c87aeb0adef7 · outbound

This paper cites TritonRL: Training LLMs to think and code triton without cheating.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation TritonRL: Training LLMs to think and code triton without cheating

Reference 38

Resolution: verified exact; arxiv_id, observed 2026-07-03T22:29:00.417504Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 13f54996-5448-4955-bbc5-6570b2ea9453 · outbound

This paper cites SWIFT: on-the-fly self-speculative decoding for LLM inference acceleration.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation SWIFT: on-the-fly self-speculative decoding for LLM inference acceleration

Reference 39

Resolution: unresolved; no resolver link, observed 2026-06-26T23:35:32.175768Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cc440eac-5a3d-41fa-bb75-cdae94597837 · outbound

This paper cites Cudaforge: An agent framework with hardware feedback for cuda kernel optimization.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation Cudaforge: An agent framework with hardware feedback for cuda kernel optimization

Reference 40

Resolution: verified exact; arxiv_id, observed 2026-07-03T22:29:00.407795Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 3301f9b6-9855-4446-98b6-219d7bfe79b2 · outbound

This paper cites Cross-attention speculative decoding.CoRR, abs/2505.24544, 2025.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation Cross-attention speculative decoding.CoRR, abs/2505.24544, 2025

Reference 41

Resolution: verified exact; arxiv_id, observed 2026-07-03T22:29:00.424238Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 2a8e9bde-9399-461c-9925-26621415a5d3 · outbound

This paper cites Qimeng-kernel: Macro-thinking micro-coding paradigm for llm-based high-performance GPU kernel generation.

SpecGen: Accelerating Agentic Kernel Optimization with Speculative Generation Qimeng-kernel: Macro-thinking micro-coding paradigm for llm-based high-performance GPU kernel generation

Reference 42

Resolution: unresolved; no resolver link, observed 2026-06-26T23:35:32.175768Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

No inbound Pith citation observations are available.