Paper Citation Record · LEDGER

Maximizing Prefix-Confidence at Test-Time Efficiently Improves Mathematical Reasoning

As of 7 August 2026, this Paper Citation Record lists 20 of 20 outbound references and 0 inbound Pith citation observations for arXiv:2507.18122.

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

pith.paper-citation-record.v1
2507.18122 v1

Coverage vector

measured 20 of 20 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-06T14:43:53.459773Z

measured 20 of 20 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

20 of 20 outbound references displayed

verified exact0
verified fuzzy4
unresolved16
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation a15ac6c0-6e00-44c6-a210-91eb0c19da9c · outbound

This paper cites The Unreasonable Effectiveness of Entropy Minimization in LLM Reasoning.

Maximizing Prefix-Confidence at Test-Time Efficiently Improves Mathematical Reasoning The Unreasonable Effectiveness of Entropy Minimization in LLM Reasoning

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-06T14:43:52.327498Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cfba05aa-50a3-4f52-af4f-a09d79aacc65 · outbound

This paper cites Training Verifiers to Solve Math Word Problems.

Maximizing Prefix-Confidence at Test-Time Efficiently Improves Mathematical Reasoning Training Verifiers to Solve Math Word Problems

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-06T14:43:52.552476Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7f73b8d6-5b7a-4b5c-b1bf-4b822c840406 · outbound

This paper cites The First Few Tokens Are All You Need: An Efficient and Effective Unsupervised Prefix Fine-Tuning Method for Reasoning Models.

Maximizing Prefix-Confidence at Test-Time Efficiently Improves Mathematical Reasoning The First Few Tokens Are All You Need: An Efficient and Effective Unsupervised Prefix Fine-Tuning Method for Reasoning Models

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-06T14:43:52.762807Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 97bdacca-4e0c-40f5-aade-522e854e6551 · outbound

This paper cites Scalable best-of-n selection for large language models via self-certainty.

Maximizing Prefix-Confidence at Test-Time Efficiently Improves Mathematical Reasoning Scalable best-of-n selection for large language models via self-certainty

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-06T14:43:52.767584Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 42b667d2-358f-497d-b686-0e1928dc395a · outbound

This paper cites This is likely because we only ever perform few gradient steps on top of the base model, and thus catastrophic forgetting does not occur.

Maximizing Prefix-Confidence at Test-Time Efficiently Improves Mathematical Reasoning This is likely because we only ever perform few gradient steps on top of the base model, and thus catastrophic forgetting does not occur

Reference 7

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T14:43:54.007293Z

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 bfc426df-37f3-4055-877d-698b692486d3 · outbound

This paper cites Confidence Is All You Need: Few-Shot RL Fine-Tuning of Language Models.

Maximizing Prefix-Confidence at Test-Time Efficiently Improves Mathematical Reasoning Confidence Is All You Need: Few-Shot RL Fine-Tuning of Language Models

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-06T14:43:52.966250Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7674c89e-55f4-45ef-843d-a13a5366a41c · outbound

This paper cites Maximizing Confidence Alone Improves Reasoning.

Maximizing Prefix-Confidence at Test-Time Efficiently Improves Mathematical Reasoning Maximizing Confidence Alone Improves Reasoning

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-06T14:43:53.050703Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c8e74a0b-eb49-4bc3-8944-0735c0ff6bcf · outbound

This paper cites Direct Reasoning Optimization: Token-Level Reasoning Reflectivity Meets Rubric Gates for Unverifiable Tasks.

Maximizing Prefix-Confidence at Test-Time Efficiently Improves Mathematical Reasoning Direct Reasoning Optimization: Token-Level Reasoning Reflectivity Meets Rubric Gates for Unverifiable Tasks

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-06T14:43:53.202824Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7b8425a9-9ec3-4a30-bb7f-dd8adec0dc74 · outbound

This paper cites Qwen2.5-Math Technical Report: Toward Mathematical Expert Model via Self-Improvement.

Maximizing Prefix-Confidence at Test-Time Efficiently Improves Mathematical Reasoning Qwen2.5-Math Technical Report: Toward Mathematical Expert Model via Self-Improvement

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-06T14:43:53.286899Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 27fc7542-72c9-4dd0-94a4-291aacb9c434 · outbound

This paper cites Absolute Zero: Reinforced Self-play Reasoning with Zero Data.

Maximizing Prefix-Confidence at Test-Time Efficiently Improves Mathematical Reasoning Absolute Zero: Reinforced Self-play Reasoning with Zero Data

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-06T14:43:53.345503Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation acdc95ec-eef2-4b19-8106-6f28a91c8166 · outbound

This paper cites Self-adapting language models.

Maximizing Prefix-Confidence at Test-Time Efficiently Improves Mathematical Reasoning Self-adapting language models

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-06T14:43:53.446619Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 31f970a2-3cf1-41c7-8c0b-f3ef5babd73a · outbound

This paper cites A Additional Confidence Measures • Negative entropy (token-level) − 1 n ∑n i=1 H(π(yi | x, y<i)) =− 1 n ∑n i=1 ∑V j=1 π(yi = j | x, y<i) log π(yi = j | x, y<i).

Maximizing Prefix-Confidence at Test-Time Efficiently Improves Mathematical Reasoning A Additional Confidence Measures • Negative entropy (token-level) − 1 n ∑n i=1 H(π(yi | x, y<i)) =− 1 n ∑n i=1 ∑V j=1 π(yi = j | x, y<i) log π(yi = j | x, y<i)

Reference 18

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T14:43:54.045748Z

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 77d7fea7-9e0b-4014-ba66-8eefbc581c99 · outbound

This paper cites Since the SGD optimizer uses less memory and compute, we continued using SGD for the rest of the experiments.

Maximizing Prefix-Confidence at Test-Time Efficiently Improves Mathematical Reasoning Since the SGD optimizer uses less memory and compute, we continued using SGD for the rest of the experiments

Reference 19

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T14:43:54.024139Z

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 d9fb6dec-c749-409a-a738-dfbb35ae1374 · outbound

This paper cites Dynamic evaluation of neural sequence models.

Maximizing Prefix-Confidence at Test-Time Efficiently Improves Mathematical Reasoning Dynamic evaluation of neural sequence models

Reference 2015

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T14:43:54.062319Z

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 8bea7096-5765-490a-b22e-fa3868e39214 · outbound

This paper cites Dynamic Evaluation of Transformer Language Models.

Maximizing Prefix-Confidence at Test-Time Efficiently Improves Mathematical Reasoning Dynamic Evaluation of Transformer Language Models

Reference 2018

Resolution: unresolved; no resolver link, observed 2026-08-06T14:43:52.841235Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 01bcdfc3-ed00-44d6-a6c4-7d5a244f1fd2 · outbound

This paper cites Self-Improving Transformers Overcome Easy-to-Hard and Length Generalization Challenges.

Maximizing Prefix-Confidence at Test-Time Efficiently Improves Mathematical Reasoning Self-Improving Transformers Overcome Easy-to-Hard and Length Generalization Challenges

Reference 2019

Resolution: unresolved; no resolver link, observed 2026-08-06T14:43:52.912793Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 56935697-518f-45f0-baba-7c650617d06e · outbound

This paper cites Learning to (Learn at Test Time): RNNs with Expressive Hidden States.

Maximizing Prefix-Confidence at Test-Time Efficiently Improves Mathematical Reasoning Learning to (Learn at Test Time): RNNs with Expressive Hidden States

Reference 2020

Resolution: unresolved; no resolver link, observed 2026-08-06T14:43:53.124912Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 792599a0-d512-40dc-aea2-acf4974531e8 · outbound

This paper cites One-Minute Video Generation with Test-Time Training.

Maximizing Prefix-Confidence at Test-Time Efficiently Improves Mathematical Reasoning One-Minute Video Generation with Test-Time Training

Reference 2021

Resolution: unresolved; no resolver link, observed 2026-08-06T14:43:52.690948Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f4ab2f1a-28d1-45fc-9b71-6c201380e3f5 · outbound

This paper cites Right Question is Already Half the Answer: Fully Unsupervised LLM Reasoning Incentivization.

Maximizing Prefix-Confidence at Test-Time Efficiently Improves Mathematical Reasoning Right Question is Already Half the Answer: Fully Unsupervised LLM Reasoning Incentivization

Reference 2022

Resolution: unresolved; no resolver link, observed 2026-08-06T14:43:53.296475Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b2ec1165-c56b-4de8-86c4-693909690b61 · outbound

This paper cites Local Mixtures of Experts: Essentially Free Test-Time Training via Model Merging.

Maximizing Prefix-Confidence at Test-Time Efficiently Improves Mathematical Reasoning Local Mixtures of Experts: Essentially Free Test-Time Training via Model Merging

Reference 2025

Resolution: unresolved; no resolver link, observed 2026-08-06T14:43:52.414007Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

No inbound Pith citation observations are available.