Paper Citation Record · LEDGER

DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

As of 4 August 2026, this Paper Citation Record lists 22 of 22 outbound references and 70 inbound Pith citation observations for arXiv:2504.11456.

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

pith.paper-citation-record.v1
2504.11456 v2

Coverage vector

measured 22 of 22 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-05-16T10:31:04.728005Z

measured 92 of 92 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-04T06:34:03.388597+00:00

measured 70 of 70 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-04T22:56:02.509530Z

measured 0 of 1 external citation measurements

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

Source: pith, observed 2026-07-09T03:25:57.840663Z

Reference resolution

22 of 22 outbound references displayed

verified exact10
verified fuzzy1
unresolved1
parse uncertain0
malformed identifier0
metadata mismatch10

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 26485edc-3072-4b24-925c-5d34a4ec5208 · outbound

This paper cites Big-Math: A Large-Scale, High-Quality Math Dataset for Reinforcement Learning in Language Models.

DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning Big-Math: A Large-Scale, High-Quality Math Dataset for Reinforcement Learning in Language Models

Reference 1

Resolution: verified exact; arxiv_id, observed 2026-05-16T10:31:04.818649Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation b931d842-000d-4550-aadb-f6fe85a48a11 · outbound

This paper cites SmolLM2: When Smol Goes Big -- Data-Centric Training of a Small Language Model.

DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning SmolLM2: When Smol Goes Big -- Data-Centric Training of a Small Language Model

Reference 2

Resolution: metadata mismatch; local_arxiv, observed 2026-05-16T10:31:04.753989Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 1689582a-6927-475c-8f48-ed77d2dfcab2 · outbound

This paper cites doi: 10.18653/v1/2023.emnlp-main.468.

DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning doi: 10.18653/v1/2023.emnlp-main.468

Reference 3

Resolution: verified exact; doi, observed 2026-05-16T10:31:04.746785Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 2d5c5678-3006-4d48-a05d-14ba7ab3a382 · outbound

This paper cites Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMs.

DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMs

Reference 4

Resolution: metadata mismatch; local_arxiv, observed 2026-05-16T10:31:04.759798Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation d4bb8b0b-0c52-479f-810f-09d643e3c0e6 · outbound

This paper cites Training Verifiers to Solve Math Word Problems.

DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning Training Verifiers to Solve Math Word Problems

Reference 5

Resolution: verified exact; local_arxiv, observed 2026-05-16T10:31:04.765546Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 46dd5051-9a31-473e-b041-bdf3603445f9 · outbound

This paper cites Process Reinforcement through Implicit Rewards.

DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning Process Reinforcement through Implicit Rewards

Reference 6

Resolution: verified exact; local_arxiv, observed 2026-05-16T10:31:04.770958Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation adfc2564-27b3-4004-9acd-8578b2f4171b · outbound

This paper cites Reinforcement learning for reasoning in small llms: What works and what doesn’t.

DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning Reinforcement learning for reasoning in small llms: What works and what doesn’t

Reference 7

Resolution: verified exact; arxiv_id, observed 2026-05-16T10:31:04.776854Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation d7fd0904-dc6b-4b76-8db5-42c463690aca · outbound

This paper cites MathOdyssey: Benchmarking Mathematical Problem-Solving Skills in Large Language Models Using Odyssey Math Data.

DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning MathOdyssey: Benchmarking Mathematical Problem-Solving Skills in Large Language Models Using Odyssey Math Data

Reference 8

Resolution: metadata mismatch; arxiv_id, observed 2026-05-16T10:31:04.783711Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 2033577b-b2b8-4cda-a571-46374563bd24 · outbound

This paper cites Cognitive Behaviors that Enable Self-Improving Reasoners, or, Four Habits of Highly Effective STaRs.

DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning Cognitive Behaviors that Enable Self-Improving Reasoners, or, Four Habits of Highly Effective STaRs

Reference 9

Resolution: verified exact; arxiv_id, observed 2026-05-17T11:40:33.380427Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 85cff00c-7d90-46d5-a7ff-42656da1cbae · outbound

This paper cites Omni-MATH: A Universal Olympiad Level Mathematic Benchmark For Large Language Models.

DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning Omni-MATH: A Universal Olympiad Level Mathematic Benchmark For Large Language Models

Reference 10

Resolution: metadata mismatch; local_arxiv, observed 2026-05-16T10:31:04.794480Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 995d756a-08af-4938-8ce5-992406dc104c · outbound

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

DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning

Reference 11

Resolution: verified exact; local_arxiv, observed 2026-05-16T10:31:04.799231Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 630c41d4-ba8c-462e-826c-c081c0fff7ab · outbound

This paper cites OlympicArena: Benchmarking Multi-discipline Cognitive Reasoning for Superintelligent AI.

DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning OlympicArena: Benchmarking Multi-discipline Cognitive Reasoning for Superintelligent AI

Reference 13

Resolution: verified exact; arxiv_id, observed 2026-05-16T10:31:04.804127Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 447177ca-6f1b-4e07-a775-d9c9af612eb1 · outbound

This paper cites an unresolved cited work.

DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning Unresolved cited work

Reference 14

Resolution: unresolved; raw_fallback, observed 2026-05-16T10:31:04.847365Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation d5894fd5-b3aa-497a-bb74-22d163c92241 · outbound

This paper cites Augmenting Math Word Problems via Iterative Question Composing.

DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning Augmenting Math Word Problems via Iterative Question Composing

Reference 15

Resolution: metadata mismatch; arxiv_id, observed 2026-05-16T10:31:04.823272Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation da5f633b-de98-4e75-bf85-fc411aeb19d8 · outbound

This paper cites AIMO-2 Winning Solution: Building State-of-the-Art Mathematical Reasoning Models with OpenMathReasoning dataset.

DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning AIMO-2 Winning Solution: Building State-of-the-Art Mathematical Reasoning Models with OpenMathReasoning dataset

Reference 16

Resolution: metadata mismatch; arxiv_id, observed 2026-05-16T10:31:04.827741Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation e3eba730-9185-4187-8c01-ad0fe37424e1 · outbound

This paper cites Sentence-bert: Sentence embeddings using siamese bert-networks.

DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning Sentence-bert: Sentence embeddings using siamese bert-networks

Reference 17

Resolution: verified fuzzy; raw_fallback, observed 2026-05-16T10:31:04.850220Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 8a257143-90ed-4c71-aeae-4cc1c5ca0954 · outbound

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

DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks

Reference 18

Resolution: metadata mismatch; local_arxiv, observed 2026-05-16T10:31:04.831503Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 8564cbcf-f582-49d1-a4be-81653f58aff0 · outbound

This paper cites DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models.

DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models

Reference 19

Resolution: metadata mismatch; local_arxiv, observed 2026-05-16T10:31:04.835825Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 5e1f518d-fd3c-4fe2-a780-f461cb2b8e55 · outbound

This paper cites OpenMathInstruct-2: Accelerating AI for Math with Massive Open-Source Instruction Data.

DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning OpenMathInstruct-2: Accelerating AI for Math with Massive Open-Source Instruction Data

Reference 20

Resolution: metadata mismatch; arxiv_id, observed 2026-05-16T10:31:04.839943Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 9cc0c66f-1531-4f89-8ac8-883264aa5fc7 · outbound

This paper cites DAPO: An Open-Source LLM Reinforcement Learning System at Scale.

DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning DAPO: An Open-Source LLM Reinforcement Learning System at Scale

Reference 21

Resolution: metadata mismatch; local_arxiv, observed 2026-05-16T10:31:04.843987Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 629dd6a8-b5e8-4559-b592-026a21ba435b · outbound

This paper cites SimpleRL-Zoo: Investigating and Taming Zero Reinforcement Learning for Open Base Models in the Wild.

DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning SimpleRL-Zoo: Investigating and Taming Zero Reinforcement Learning for Open Base Models in the Wild

Reference 22

Resolution: verified exact; local_arxiv, observed 2026-05-16T10:31:04.813214Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation ae8240d8-0d89-450b-876c-8366c45afb75 · outbound

This paper cites MegaMath: Pushing the Limits of Open Math Corpora.

DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning MegaMath: Pushing the Limits of Open Math Corpora

Reference 23

Resolution: verified exact; arxiv_id, observed 2026-05-16T10:31:04.809020Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Pith citing papers

Observation 47bbb0d6-6ebb-4c6e-a885-d781984ddf9a · inbound

OpenVLThinker: Complex Vision-Language Reasoning via Iterative SFT-RL Cycles cites this paper.

OpenVLThinker: Complex Vision-Language Reasoning via Iterative SFT-RL Cycles DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 23

Resolution: verified exact; local_arxiv, observed 2026-05-19T06:59:03.404797Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 058bade5-4ca3-4d40-aae4-2ec52444afe0 · inbound

Beyond Two-Stage Training: Cooperative SFT and RL for LLM Reasoning cites this paper.

Beyond Two-Stage Training: Cooperative SFT and RL for LLM Reasoning DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-04T22:56:02.509530Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a60124cb-8aa4-426d-9994-f1153aa919bd · inbound

A Survey of Reinforcement Learning for Large Reasoning Models cites this paper.

A Survey of Reinforcement Learning for Large Reasoning Models DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 193

Resolution: verified exact; local_arxiv, observed 2026-05-18T00:05:31.546896Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 723679c4-7e2d-4f29-aa74-cb04aaf5aadf · inbound

Plan Then Action:High-Level Planning Guidance Reinforcement Learning for LLM Reasoning cites this paper.

Plan Then Action:High-Level Planning Guidance Reinforcement Learning for LLM Reasoning DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-04T12:52:26.866176Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation efedbebc-b4da-49ff-8023-bf8b809b4272 · inbound

Probing the Difficulty Perception Mechanism of Large Language Models cites this paper.

Probing the Difficulty Perception Mechanism of Large Language Models DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-04T11:17:47.562551Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 967a59f4-c332-409d-a416-48d6fa20ece1 · inbound

Rewarding Structural Conformance of Reasoning using Process Mining cites this paper.

Rewarding Structural Conformance of Reasoning using Process Mining DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-04T07:37:16.657252Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7eff8bab-4ce0-4f8b-b813-c430af938f21 · inbound

RLVE: Scaling Up Reinforcement Learning for Language Models with Adaptive Verifiable Environments cites this paper.

RLVE: Scaling Up Reinforcement Learning for Language Models with Adaptive Verifiable Environments DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-03T23:08:05.560042Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5e0f3afa-9e22-4072-a047-c2e47d2f7927 · inbound

SPHINX: A Synthetic Environment for Visual Perception and Reasoning cites this paper.

SPHINX: A Synthetic Environment for Visual Perception and Reasoning DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 20

Resolution: verified exact; local_arxiv, observed 2026-05-17T04:21:30.768641Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation c73c02c1-8d22-4ee2-bebf-4d4972f38c20 · inbound

SCALER:Synthetic Scalable Adaptive Learning Environment for Reasoning cites this paper.

SCALER:Synthetic Scalable Adaptive Learning Environment for Reasoning DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 14

Resolution: verified exact; local_arxiv, observed 2026-05-16T16:28:05.679417Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation df99729a-bc3a-4ae8-9baa-f3fa3296fdb8 · inbound

Conversation for Non-verifiable Learning: Self-Evolving LLMs through Meta-Evaluation cites this paper.

Conversation for Non-verifiable Learning: Self-Evolving LLMs through Meta-Evaluation DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 9

Resolution: verified exact; arxiv_id, observed 2026-05-16T10:31:04.851073Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation cf38665a-e33a-400c-b7a5-84870c2e7f42 · inbound

On the Overscaling Curse of Parallel Thinking: System Efficacy Contradicts Sample Efficiency cites this paper.

On the Overscaling Curse of Parallel Thinking: System Efficacy Contradicts Sample Efficiency DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 11

Resolution: verified exact; local_arxiv, observed 2026-05-16T10:40:51.353235Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 860ebc19-c46d-4c0c-8fb0-6d1bd6a6f7f8 · inbound

GradAlign: Gradient-Aligned Data Selection for LLM Reinforcement Learning cites this paper.

GradAlign: Gradient-Aligned Data Selection for LLM Reinforcement Learning DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-02T21:05:25.136603Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f9d9e73e-2c42-43b6-88a5-c55fc99f6445 · inbound

Compress the Easy, Explore the Hard: Difficulty-Aware Entropy Regularization for Efficient LLM Reasoning cites this paper.

Compress the Easy, Explore the Hard: Difficulty-Aware Entropy Regularization for Efficient LLM Reasoning DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-02T20:44:44.881935Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 690dc376-9abc-4168-8dd7-8c8e5a7cf793 · inbound

Your Model Diversity, Not Method, Determines Reasoning Strategy cites this paper.

Your Model Diversity, Not Method, Determines Reasoning Strategy DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 4

Resolution: verified exact; arxiv_id, observed 2026-05-16T10:31:04.851073Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 07c5dbba-d228-4d05-aeae-fa4ae0032d28 · inbound

PubSwap: Public-Data Off-Policy Coordination for Federated RLVR cites this paper.

PubSwap: Public-Data Off-Policy Coordination for Federated RLVR DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 4

Resolution: verified exact; arxiv_id, observed 2026-05-16T10:31:04.851073Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation c4d001c6-05c1-4863-a9cb-572442a58a36 · inbound

Rethinking the Comparison Unit in Sequence-Level Reinforcement Learning: An Equal-Length Paired Training Framework from Loss Correction to Sample Construction cites this paper.

Rethinking the Comparison Unit in Sequence-Level Reinforcement Learning: An Equal-Length Paired Training Framework from Loss Correction to Sample Construction DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 25

Resolution: verified exact; arxiv_id, observed 2026-05-16T10:31:04.851073Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 94fced4b-444c-4351-a118-0b4e0b12381f · inbound

Learning from Less: Measuring the Effectiveness of RLVR in Low Data and Compute Regimes cites this paper.

Learning from Less: Measuring the Effectiveness of RLVR in Low Data and Compute Regimes DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 5

Resolution: verified exact; arxiv_id, observed 2026-05-16T10:31:04.851073Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 0d3b114b-7072-4f2c-83fe-8eb9bab0386a · inbound

Too Correct to Learn: Reinforcement Learning on Saturated Reasoning Data cites this paper.

Too Correct to Learn: Reinforcement Learning on Saturated Reasoning Data DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 73

Resolution: metadata mismatch; arxiv_id, observed 2026-05-16T10:31:04.851073Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 8997fca0-e118-4973-a178-c56d88e2a3c4 · inbound

Length Value Model: Scalable Value Pretraining for Token-Level Length Modeling cites this paper.

Length Value Model: Scalable Value Pretraining for Token-Level Length Modeling DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 8

Resolution: verified exact; arxiv_id, observed 2026-05-16T10:31:04.851073Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 3e7b28f8-4b77-43c5-b8f0-cb93f6526045 · inbound

Length Value Model: Scalable Value Pretraining for Token-Level Length Modeling cites this paper.

Length Value Model: Scalable Value Pretraining for Token-Level Length Modeling DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-02T15:28:11.698960Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 84e34f6c-4458-468d-9d27-fca1959a941f · inbound

Internalizing Outcome Supervision into Process Supervision: A New Paradigm for Reinforcement Learning for Reasoning cites this paper.

Internalizing Outcome Supervision into Process Supervision: A New Paradigm for Reinforcement Learning for Reasoning DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 29

Resolution: verified exact; arxiv_id, observed 2026-05-16T10:31:04.851073Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation bfebc584-9595-40c8-a585-a60125dd2e88 · inbound

Asymmetric On-Policy Distillation: Bridging Exploitation and Imitation at the Token Level cites this paper.

Asymmetric On-Policy Distillation: Bridging Exploitation and Imitation at the Token Level DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 19

Resolution: metadata mismatch; arxiv_id, observed 2026-05-16T10:31:04.851073Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 32a3b2f2-5629-4d01-ab47-dc53fee74265 · inbound

Asymmetric On-Policy Distillation: Bridging Exploitation and Imitation at the Token Level cites this paper.

Reference 19

Resolution: metadata mismatch; arxiv_id, observed 2026-05-16T10:31:04.851073Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 0515b3fd-630d-4969-bf5d-2338303bdbb1 · inbound

Asymmetric On-Policy Distillation: Bridging Exploitation and Imitation at the Token Level cites this paper.

Reference 19

Resolution: metadata mismatch; arxiv_id, observed 2026-05-16T10:31:04.851073Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation efa33de1-b97f-4433-a238-25112f42335b · inbound

On the Implicit Reward Overfitting and the Low-rank Dynamics in RLVR cites this paper.

On the Implicit Reward Overfitting and the Low-rank Dynamics in RLVR DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 12

Resolution: verified exact; arxiv_id, observed 2026-05-16T10:31:04.851073Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 685b4bac-0184-4bf8-99b8-1ceab8937d93 · inbound

AIPO: Learning to Reason from Active Interaction cites this paper.

AIPO: Learning to Reason from Active Interaction DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 21

Resolution: verified exact; arxiv_id, observed 2026-05-16T10:31:04.851073Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 47286ca3-2199-4373-9bf6-b33efca7cdb3 · inbound

AIPO: Learning to Reason from Active Interaction cites this paper.

AIPO: Learning to Reason from Active Interaction DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 21

Resolution: verified exact; local_arxiv, observed 2026-05-19T18:07:42.397935Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation b35cf74e-1a9d-4ee5-ada3-20189188e3c4 · inbound

Mid-Training with Self-Generated Data Improves Reinforcement Learning in Language Models cites this paper.

Mid-Training with Self-Generated Data Improves Reinforcement Learning in Language Models DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 16

Resolution: verified exact; arxiv_id, observed 2026-05-16T10:31:04.851073Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 894a6985-5629-4100-99a9-42a827d32dfa · inbound

Training with Harnesses: On-Policy Harness Self-Distillation for Complex Reasoning cites this paper.

Training with Harnesses: On-Policy Harness Self-Distillation for Complex Reasoning DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 14

Resolution: verified exact; arxiv_id, observed 2026-05-16T10:31:04.851073Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 7472f34e-0af5-472c-885f-591472bf5d2a · inbound

Forge: Quality-Aware Reinforcement Learning for NP-Hard Optimization in LLMs cites this paper.

Forge: Quality-Aware Reinforcement Learning for NP-Hard Optimization in LLMs DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 17

Resolution: metadata mismatch; arxiv_id, observed 2026-05-16T10:31:04.851073Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 22d33486-cbbf-4c6a-95c9-20ae25de4097 · inbound

DARE: Difficulty-Adaptive Reinforcement Learning with Co-Evolved Difficulty Estimation cites this paper.

DARE: Difficulty-Adaptive Reinforcement Learning with Co-Evolved Difficulty Estimation DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 13

Resolution: verified exact; arxiv_id, observed 2026-05-16T10:31:04.851073Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation e0eb2663-a354-47ec-91a2-097eb8ad1ec1 · inbound

Nice Fold or Hero Call: Learning Budget-Efficient Thinking for Adaptive Reasoning cites this paper.

Nice Fold or Hero Call: Learning Budget-Efficient Thinking for Adaptive Reasoning DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 18

Resolution: verified exact; arxiv_id, observed 2026-05-16T10:31:04.851073Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 63e4046a-1b0f-43a9-99ce-44fb76fbc868 · inbound

Learning to Foresee: Unveiling the Unlocking Efficiency of On-Policy Distillation cites this paper.

Learning to Foresee: Unveiling the Unlocking Efficiency of On-Policy Distillation DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 11

Resolution: verified exact; local_arxiv, observed 2026-05-22T10:11:22.765572Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 23eeed74-71e7-412f-b5eb-958a2afb8867 · inbound

Pion: A Spectrum-Preserving Optimizer via Orthogonal Equivalence Transformation cites this paper.

Pion: A Spectrum-Preserving Optimizer via Orthogonal Equivalence Transformation DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 30

Resolution: verified exact; arxiv_id, observed 2026-05-16T10:31:04.851073Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation ad9b5f8d-ce88-4956-b600-102f09a6f772 · inbound

Multi-Rollout On-Policy Distillation via Peer Successes and Failures cites this paper.

Multi-Rollout On-Policy Distillation via Peer Successes and Failures DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 73

Resolution: metadata mismatch; arxiv_id, observed 2026-05-16T10:31:04.851073Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 550971ca-a461-4aae-96f8-1821c3e2c580 · inbound

Taming the Thinker: Conditional Entropy Shaping for Adaptive LLM Reasoning cites this paper.

Taming the Thinker: Conditional Entropy Shaping for Adaptive LLM Reasoning DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 23

Resolution: metadata mismatch; local_arxiv, observed 2026-05-20T06:43:05.906903Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation e4b2e5c6-750a-41d0-b97e-86ef77a03cc4 · inbound

Beyond Mode Collapse: Distribution Matching for Diverse Reasoning cites this paper.

Beyond Mode Collapse: Distribution Matching for Diverse Reasoning DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 15

Resolution: verified exact; local_arxiv, observed 2026-05-20T05:33:04.086260Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 971dcc5f-8264-42b0-8b6c-464b61374a60 · inbound

DEL: Digit Entropy Loss for Numerical Learning of Large Language Models cites this paper.

DEL: Digit Entropy Loss for Numerical Learning of Large Language Models DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 52

Resolution: verified exact; local_arxiv, observed 2026-05-21T07:39:49.460219Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation bd7ea9ff-da6f-420d-bdd9-6ee31bad6352 · inbound

DelTA: Discriminative Token Credit Assignment for Reinforcement Learning from Verifiable Rewards cites this paper.

DelTA: Discriminative Token Credit Assignment for Reinforcement Learning from Verifiable Rewards DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 54

Resolution: verified exact; local_arxiv, observed 2026-05-21T05:29:40.132496Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 8bb3ca51-c57b-4648-895e-466727ea8ea2 · inbound

LANG: Reinforcement Learning for Multilingual Reasoning with Language-Adaptive Hint Guidance cites this paper.

LANG: Reinforcement Learning for Multilingual Reasoning with Language-Adaptive Hint Guidance DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 31

Resolution: metadata mismatch; local_arxiv, observed 2026-05-22T06:21:10.412848Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation e880cc08-4db1-45fc-9a62-cd7c46fb811b · inbound

Not only where, But when: Temporal Scheduling for RLVR cites this paper.

Not only where, But when: Temporal Scheduling for RLVR DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 13

Resolution: verified exact; local_arxiv, observed 2026-06-29T22:44:01.863471Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation ea838eab-8e97-476f-8d80-a2736669e7f2 · inbound

Spend Your Rollouts Where It Counts: Rollout Allocation for Group-Based RL Post-Training cites this paper.

Spend Your Rollouts Where It Counts: Rollout Allocation for Group-Based RL Post-Training DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 4

Resolution: verified exact; local_arxiv, observed 2026-06-29T19:53:55.807818Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation bda730de-94b9-4c7c-8f93-9450c7624124 · inbound

RLVR Datasets and Where to Find Them: Tracing Data Lineage for Better Training Data cites this paper.

RLVR Datasets and Where to Find Them: Tracing Data Lineage for Better Training Data DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 16

Resolution: verified exact; local_arxiv, observed 2026-06-29T19:43:55.115046Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 9bee5b38-0f96-4678-a3e3-9faac621eb5c · inbound

DenoiseRL: Bootstrapping Reasoning Models to Recover from Noisy Prefixes cites this paper.

DenoiseRL: Bootstrapping Reasoning Models to Recover from Noisy Prefixes DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 9

Resolution: verified exact; local_arxiv, observed 2026-06-29T11:53:23.707364Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation d7258fbd-1837-4aa7-8859-85c5e3b47204 · inbound

DenoiseRL: Bootstrapping Reasoning Models to Recover from Noisy Prefixes cites this paper.

DenoiseRL: Bootstrapping Reasoning Models to Recover from Noisy Prefixes DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-02T12:58:10.897147Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a769edab-07f2-4b0d-9d8a-e2a919c54835 · inbound

Trust Region On-Policy Distillation cites this paper.

Trust Region On-Policy Distillation DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 287

Resolution: metadata mismatch; local_arxiv, observed 2026-07-01T20:56:13.352648Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation a757bd81-2922-43b1-bea3-55363c01d36a · inbound

Filter, Then Reweight: Rethinking Optimization Granularity in On-Policy Distillation cites this paper.

Filter, Then Reweight: Rethinking Optimization Granularity in On-Policy Distillation DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 22

Resolution: metadata mismatch; local_arxiv, observed 2026-07-01T22:36:17.403843Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 7fb7b005-ff50-4cc2-a597-31b394148e56 · inbound

ThoughtFold: Folding Reasoning Chains via Introspective Preference Learning cites this paper.

ThoughtFold: Folding Reasoning Chains via Introspective Preference Learning DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 11

Resolution: metadata mismatch; local_arxiv, observed 2026-07-02T03:26:28.773294Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation c980f253-16a5-419b-9818-56453ac27ce8 · inbound

GeoMin: Data-Efficient Semi-Supervised RLVR via Geometric Distribution Modeling cites this paper.

GeoMin: Data-Efficient Semi-Supervised RLVR via Geometric Distribution Modeling DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 26

Resolution: metadata mismatch; local_arxiv, observed 2026-07-02T06:06:40.776163Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 9a80c423-7e11-43b9-8251-bb8b81bd5859 · inbound

GRAIL: Gradient-Reweighted Advantages for Reinforcement Learning with Verifiable Rewards cites this paper.

GRAIL: Gradient-Reweighted Advantages for Reinforcement Learning with Verifiable Rewards DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 6

Resolution: verified exact; local_arxiv, observed 2026-07-02T08:36:47.619392Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation b425e865-08fb-4d09-aafc-324d5546c6da · inbound

State commitment learning: training language models to distinguish computation from memory cites this paper.

State commitment learning: training language models to distinguish computation from memory DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 5

Resolution: verified exact; local_arxiv, observed 2026-06-30T15:24:49.920750Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 457f3a84-162b-4381-8e5e-f7a65fda66f9 · inbound

ConSteer-RL: Steering Reasoning Capabilities in Large Language Models via Confidence-Aware Reinforcement Learning cites this paper.

ConSteer-RL: Steering Reasoning Capabilities in Large Language Models via Confidence-Aware Reinforcement Learning DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 2

Resolution: metadata mismatch; local_arxiv, observed 2026-07-02T21:07:23.905688Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 8249fa7b-1492-4cc3-a41a-584ac4c82c17 · inbound

SG-OPD: Sign-Gated On-Policy Distillation via Sign-Consistency Gating and Phased Teacher Sampling cites this paper.

SG-OPD: Sign-Gated On-Policy Distillation via Sign-Consistency Gating and Phased Teacher Sampling DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 1

Resolution: metadata mismatch; local_arxiv, observed 2026-07-03T01:17:31.317061Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 8c83603f-2553-43f7-9c8a-8d0ba9286330 · inbound

RLCSD: Reinforcement Learning with Contrastive On-Policy Self-Distillation cites this paper.

RLCSD: Reinforcement Learning with Contrastive On-Policy Self-Distillation DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 8

Resolution: verified exact; local_arxiv, observed 2026-07-03T09:07:48.370451Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 365dd4e4-44fb-4a6a-980e-69e8631d0689 · inbound

Shattering the Autoregressive Curse: Dynamic Epistemic Entropy Orchestrated Erasable Reinforcement Learning for LLMs cites this paper.

Shattering the Autoregressive Curse: Dynamic Epistemic Entropy Orchestrated Erasable Reinforcement Learning for LLMs DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 33

Resolution: verified exact; local_arxiv, observed 2026-07-03T21:18:58.389882Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 74c5ef6d-7dda-442c-b49a-f823e2885d23 · inbound

Finding the Evidence: Discovering Decision-Supporting Tokens for On-Policy Reasoning Distillation cites this paper.

Finding the Evidence: Discovering Decision-Supporting Tokens for On-Policy Reasoning Distillation DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 8

Resolution: metadata mismatch; local_arxiv, observed 2026-07-04T10:29:44.952266Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 1cc6cde3-6c36-4222-87cd-4a0aa0896f7b · inbound

AsyncOPD: How Stale Can On-Policy Distillation Be? cites this paper.

AsyncOPD: How Stale Can On-Policy Distillation Be? DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 8

Resolution: verified exact; local_arxiv, observed 2026-07-04T16:29:56.821408Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 2832645c-535e-4fe9-8480-08a3aa464086 · inbound

EntroRouter: Learning Efficient Model Routing via Entropy Regulation cites this paper.

EntroRouter: Learning Efficient Model Routing via Entropy Regulation DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 45

Resolution: metadata mismatch; local_arxiv, observed 2026-06-30T07:44:21.623493Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation ea29cd36-3041-4a6c-aa58-ad151fe1e294 · inbound

The Mirage of Optimizing Training Policies: Monotonic Inference Policies as the Real Objective for LLM Reinforcement Learning cites this paper.

The Mirage of Optimizing Training Policies: Monotonic Inference Policies as the Real Objective for LLM Reinforcement Learning DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 2

Resolution: metadata mismatch; local_arxiv, observed 2026-06-30T07:24:20.891805Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 0cd46127-b12a-4cf3-9c6a-f76ce529268f · inbound

Beyond Compilation: Evaluating Faithful Natural-Language-to-Lean Statement Formalization cites this paper.

Beyond Compilation: Evaluating Faithful Natural-Language-to-Lean Statement Formalization DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 19

Resolution: metadata mismatch; local_arxiv, observed 2026-07-01T13:15:45.431035Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation f640058f-4066-40c0-b8b7-8541b8dd9730 · inbound

On the effectiveness of reward functions in reinforcement learning for confidence calibration of large language models cites this paper.

On the effectiveness of reward functions in reinforcement learning for confidence calibration of large language models DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 22

Resolution: unresolved; no resolver link, observed 2026-07-11T20:01:06.623978Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 78c2130d-9121-4dd8-9f88-afced1b0401f · inbound

Weak-to-Strong Generalization via Direct On-Policy Distillation cites this paper.

Weak-to-Strong Generalization via Direct On-Policy Distillation DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 16

Resolution: verified exact; local_arxiv, observed 2026-07-07T12:33:45.202425Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 6cf49e12-3bfb-4301-9504-f6cef34cf187 · inbound

Weak-to-Strong Generalization via Direct On-Policy Distillation cites this paper.

Weak-to-Strong Generalization via Direct On-Policy Distillation DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 16

Resolution: unresolved; no resolver link, observed 2026-07-11T07:01:56.628017Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2dea5867-8b5f-49dc-9414-306428d24929 · inbound

Max Out GRPO Signal: Adaptive Trace Prefix Control for Hard Reasoning Problems cites this paper.

Max Out GRPO Signal: Adaptive Trace Prefix Control for Hard Reasoning Problems DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 19

Resolution: verified exact; local_arxiv, observed 2026-07-09T03:25:57.841864Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 8b390d5b-bf7a-42a2-b48e-ce75570111dc · inbound

Agon: Competitive Cross-Model RL with Implicit Rival Grading of Reasoning cites this paper.

Agon: Competitive Cross-Model RL with Implicit Rival Grading of Reasoning DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 1

Resolution: verified exact; local_arxiv, observed 2026-07-09T02:35:53.868114Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 8c12d294-4409-4f4e-a72f-2f576ca10288 · inbound

Proxy Exploration and Reusable Guidance: A Modular LLM Post-Training Paradigm via Proxy-Guided Update Signals cites this paper.

Proxy Exploration and Reusable Guidance: A Modular LLM Post-Training Paradigm via Proxy-Guided Update Signals DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 18

Resolution: unresolved; no resolver link, observed 2026-07-14T05:09:00.865375Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3b5f5768-87a2-432a-88c7-92682d12d1ef · inbound

Cross-Tokenizer On-Policy Distillation via Byte-Prefix Marginalization cites this paper.

Cross-Tokenizer On-Policy Distillation via Byte-Prefix Marginalization DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 59

Resolution: unresolved; no resolver link, observed 2026-08-01T05:09:58.972025Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3c36225b-6de3-4a63-8363-09d4bb8883ad · inbound

Weak-to-Strong On-Policy Distillation cites this paper.

Weak-to-Strong On-Policy Distillation DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 76

Resolution: unresolved; no resolver link, observed 2026-08-01T00:26:24.869605Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 72fd49f8-e1e1-40ae-961e-20159b36095b · inbound

LoRA Scaffolded Policy Optimization (LSPO): A Sampling-Time Low-Rank Scaffold for Recovering Reinforcement-Learning Gradient on Zero-Reward Cliff Prompts cites this paper.

LoRA Scaffolded Policy Optimization (LSPO): A Sampling-Time Low-Rank Scaffold for Recovering Reinforcement-Learning Gradient on Zero-Reward Cliff Prompts DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-01T01:28:56.250910Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b63273ff-1c3c-404e-80a4-b52983f59eaf · inbound

ThinkReset: Learnable Intermediate Interface Construction for Bounded-Context Long-Horizon Reasoning cites this paper.

ThinkReset: Learnable Intermediate Interface Construction for Bounded-Context Long-Horizon Reasoning DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-03T00:54:34.691461Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.