Paper Citation Record · LEDGER

OPTScientist: Multi-Agent Discovery of Typed Optimizer Programs for Transformer Pretraining

As of 14 August 2026, this Paper Citation Record lists 19 of 19 outbound references and 0 inbound Pith citation observations for arXiv:2607.20486.

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

pith.paper-citation-record.v1
2607.20486 v1

Coverage vector

measured 19 of 19 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-02T12:30:38.516376Z

measured 19 of 19 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-14T06:32:32.682623+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

19 of 19 outbound references displayed

verified exact0
verified fuzzy0
unresolved19
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation f6dcbe8f-a36d-41d6-8e75-719a87c5001c · outbound

This paper cites Muon is Scalable for LLM Training.

OPTScientist: Multi-Agent Discovery of Typed Optimizer Programs for Transformer Pretraining Muon is Scalable for LLM Training

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-02T12:30:37.205102Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 022d59af-d086-4986-93d6-aa36d5f876cc · outbound

This paper cites Evolving deep learning optimizers.arXiv preprint arXiv:2512.11853,.

OPTScientist: Multi-Agent Discovery of Typed Optimizer Programs for Transformer Pretraining Evolving deep learning optimizers.arXiv preprint arXiv:2512.11853,

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-02T12:30:37.406691Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0b9ff42a-70c5-4c97-9992-36fc451b415f · outbound

This paper cites Celo: Training Versatile Learned Optimizers on a Compute Diet.

OPTScientist: Multi-Agent Discovery of Typed Optimizer Programs for Transformer Pretraining Celo: Training Versatile Learned Optimizers on a Compute Diet

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-02T12:30:37.510251Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b7211e0b-b8ee-438c-a80a-21aa0ab0b049 · outbound

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

OPTScientist: Multi-Agent Discovery of Typed Optimizer Programs for Transformer Pretraining AlphaEvolve: A coding agent for scientific and algorithmic discovery

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-02T12:30:37.570409Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d082cccb-7185-4a1f-b115-e979fe04e62c · outbound

This paper cites The AdEMAMix Optimizer: Better, Faster, Older.

OPTScientist: Multi-Agent Discovery of Typed Optimizer Programs for Transformer Pretraining The AdEMAMix Optimizer: Better, Faster, Older

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-02T12:30:37.647532Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b3646b4c-f5df-4c5c-8f89-1eebbad837e2 · outbound

This paper cites Agent laboratory: Using llm agents as research assistants.Findings of the Association for Computational Linguistics: EMNLP 2025, pages 5977–6043,.

OPTScientist: Multi-Agent Discovery of Typed Optimizer Programs for Transformer Pretraining Agent laboratory: Using llm agents as research assistants.Findings of the Association for Computational Linguistics: EMNLP 2025, pages 5977–6043,

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-02T12:30:37.769775Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1e0fbc28-13f5-4a0c-a40b-e1eed22d1611 · outbound

This paper cites SOAP: Improving and Stabilizing Shampoo using Adam.

OPTScientist: Multi-Agent Discovery of Typed Optimizer Programs for Transformer Pretraining SOAP: Improving and Stabilizing Shampoo using Adam

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-02T12:30:37.840921Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d3f0ab87-0d9a-43d4-b28e-24ce0c57e9ee · outbound

This paper cites Voyager: An Open-Ended Embodied Agent with Large Language Models.

OPTScientist: Multi-Agent Discovery of Typed Optimizer Programs for Transformer Pretraining Voyager: An Open-Ended Embodied Agent with Large Language Models

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-02T12:30:37.928953Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b463f58c-5e27-4376-9c2e-65ce27c0606e · outbound

This paper cites Fantastic Pretraining Optimizers and Where to Find Them.

OPTScientist: Multi-Agent Discovery of Typed Optimizer Programs for Transformer Pretraining Fantastic Pretraining Optimizers and Where to Find Them

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-02T12:30:38.010996Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 57ec6676-3b01-43eb-a35f-c398c4a2b9ae · outbound

This paper cites ReAct: Synergizing Reasoning and Acting in Language Models.

OPTScientist: Multi-Agent Discovery of Typed Optimizer Programs for Transformer Pretraining ReAct: Synergizing Reasoning and Acting in Language Models

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-02T12:30:38.113503Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0f32997d-8304-4d9a-9fa0-e4c2397f32b3 · outbound

This paper cites MARS: Unleashing the Power of Variance Reduction for Training Large Models.

OPTScientist: Multi-Agent Discovery of Typed Optimizer Programs for Transformer Pretraining MARS: Unleashing the Power of Variance Reduction for Training Large Models

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-02T12:30:38.237293Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a1d73cfc-f921-4284-ba34-3986618c46ee · outbound

This paper cites Adam-mini: Use Fewer Learning Rates To Gain More.

OPTScientist: Multi-Agent Discovery of Typed Optimizer Programs for Transformer Pretraining Adam-mini: Use Fewer Learning Rates To Gain More

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-02T12:30:38.316916Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9ab70c78-1040-4bde-8430-3410abfde270 · outbound

This paper cites GaLore: Memory-Efficient LLM Training by Gradient Low-Rank Projection.

OPTScientist: Multi-Agent Discovery of Typed Optimizer Programs for Transformer Pretraining GaLore: Memory-Efficient LLM Training by Gradient Low-Rank Projection

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-02T12:30:38.429120Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 42760fe0-916a-4f08-80f8-ccd4e16cb9d1 · outbound

This paper cites The main paper only introduces the high-level idea: optimizers are represented as modular update programs.

OPTScientist: Multi-Agent Discovery of Typed Optimizer Programs for Transformer Pretraining The main paper only introduces the high-level idea: optimizers are represented as modular update programs

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-02T12:30:38.516376Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 10073dda-28f5-4d69-b5fb-6182f31873af · outbound

This paper cites Researchagent: Iterative research idea generation over scientific literature with large language models.

OPTScientist: Multi-Agent Discovery of Typed Optimizer Programs for Transformer Pretraining Researchagent: Iterative research idea generation over scientific literature with large language models

Reference 2016

Resolution: unresolved; no resolver link, observed 2026-08-02T12:30:36.865277Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 28c76fd8-8bc7-4bd6-aa2a-b901780789a1 · outbound

This paper cites The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery.

OPTScientist: Multi-Agent Discovery of Typed Optimizer Programs for Transformer Pretraining The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery

Reference 2019

Resolution: unresolved; no resolver link, observed 2026-08-02T12:30:37.293148Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 962eaa74-353e-4565-a227-bdb8082bf18a · outbound

This paper cites Evolution of Heuristics: Towards Efficient Automatic Algorithm Design Using Large Language Model.

OPTScientist: Multi-Agent Discovery of Typed Optimizer Programs for Transformer Pretraining Evolution of Heuristics: Towards Efficient Automatic Algorithm Design Using Large Language Model

Reference 2023

Resolution: unresolved; no resolver link, observed 2026-08-02T12:30:37.006708Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a0823259-c6e0-43b6-be8a-72bef1918ab1 · outbound

This paper cites Sophia: A Scalable Stochastic Second-order Optimizer for Language Model Pre-training.

OPTScientist: Multi-Agent Discovery of Typed Optimizer Programs for Transformer Pretraining Sophia: A Scalable Stochastic Second-order Optimizer for Language Model Pre-training

Reference 2024

Resolution: unresolved; no resolver link, observed 2026-08-02T12:30:37.114754Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c5f72569-e566-4879-8ea1-5b60ad5cbeac · outbound

This paper cites Adam: A Method for Stochastic Optimization.

OPTScientist: Multi-Agent Discovery of Typed Optimizer Programs for Transformer Pretraining Adam: A Method for Stochastic Optimization

Reference 2025

Resolution: unresolved; no resolver link, observed 2026-08-02T12:30:36.923611Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

No inbound Pith citation observations are available.