Paper Citation Record · LEDGER

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability

As of 17 August 2026, this Paper Citation Record lists 27 of 27 outbound references and 27 inbound Pith citation observations for arXiv:2411.19943.

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

pith.paper-citation-record.v1
2411.19943 v3

Coverage vector

measured 27 of 27 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-12T05:43:32.296269Z

measured 54 of 54 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-17T06:30:58.91139+00:00

measured 27 of 27 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-09T19:37:15.914764Z

measured 0 of 1 external citation measurements

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

Source: pith, observed 2026-07-08T01:14:27.537549Z

Reference resolution

27 of 27 outbound references displayed

verified exact1
verified fuzzy1
unresolved25
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation cff932a3-4fae-44d8-a0a8-e6e0e818b577 · outbound

This paper cites Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback.

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-12T05:43:31.261632Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b3f33ca0-e460-4c96-a9c3-07cab5877c59 · outbound

This paper cites Adversarial Contrastive Estimation.

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability Adversarial Contrastive Estimation

Reference 4

Resolution: verified exact; local_arxiv, observed 2026-08-12T05:43:33.138292Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 2e3a336b-d35e-412a-a212-196ae26a005e · outbound

This paper cites Towards Analyzing and Understanding the Limitations of DPO: A Theoretical Perspective.

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability Towards Analyzing and Understanding the Limitations of DPO: A Theoretical Perspective

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-12T05:43:31.415705Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7af01e0a-c023-4038-90e9-ac3e09839277 · outbound

This paper cites Chain-of-Thought Hub: A Continuous Effort to Measure Large Language Models' Reasoning Performance.

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability Chain-of-Thought Hub: A Continuous Effort to Measure Large Language Models' Reasoning Performance

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-12T05:43:31.508571Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9b99a00e-8979-476f-8f7d-adafd4504621 · outbound

This paper cites Beyond Imitation: Leveraging Fine-grained Quality Signals for Alignment.

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability Beyond Imitation: Leveraging Fine-grained Quality Signals for Alignment

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-12T05:43:31.514283Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 232c31f4-a495-4526-85c3-39f190e2e66e · outbound

This paper cites Step-DPO: Step-wise Preference Optimization for Long-chain Reasoning of LLMs.

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability Step-DPO: Step-wise Preference Optimization for Long-chain Reasoning of LLMs

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-12T05:43:31.521431Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6dc7fcab-5400-449c-a40a-90a3ecd5c894 · outbound

This paper cites RewardBench: Evaluating Reward Models for Language Modeling.

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability RewardBench: Evaluating Reward Models for Language Modeling

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-12T05:43:31.526928Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2c6fef85-2a72-43d5-aa2d-e7a57d611b01 · outbound

This paper cites Let's Verify Step by Step.

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability Let's Verify Step by Step

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-12T05:43:31.531366Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f421e724-8314-48ab-9ee7-d48c232a2812 · outbound

This paper cites Provably Mitigating Overoptimization in RLHF: Your SFT Loss is Implicitly an Adversarial Regularizer.

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability Provably Mitigating Overoptimization in RLHF: Your SFT Loss is Implicitly an Adversarial Regularizer

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-12T05:43:31.667639Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation acf1ed0b-174e-4538-a19d-d232d546d671 · outbound

This paper cites Contrastive Decoding Improves Reasoning in Large Language Models.

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability Contrastive Decoding Improves Reasoning in Large Language Models

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-12T05:43:31.769301Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6cbd74b3-9cbb-4d9f-9654-84612b71472b · outbound

This paper cites Smaug: Fixing Failure Modes of Preference Optimisation with DPO-Positive.

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability Smaug: Fixing Failure Modes of Preference Optimisation with DPO-Positive

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-12T05:43:31.785406Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a78b8149-e8ae-4083-94a3-7d9068f26c9c · outbound

This paper cites Iterative Reasoning Preference Optimization.

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability Iterative Reasoning Preference Optimization

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-12T05:43:31.793155Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 17f4f9bd-254e-4139-8cfd-04f87749d09b · outbound

This paper cites Proximal Policy Optimization Algorithms.

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability Proximal Policy Optimization Algorithms

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-12T05:43:31.798231Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3bc083fb-f832-4448-b9f6-6aa3fbb794b1 · outbound

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

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-12T05:43:31.803645Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 41207014-57b7-437e-a048-0c6030066147 · outbound

This paper cites Unchosen Experts Can Contribute Too: Unleashing MoE Models' Power by Self-Contrast.

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability Unchosen Experts Can Contribute Too: Unleashing MoE Models' Power by Self-Contrast

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-12T05:43:31.926793Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6ee4dcdc-1f38-4535-82d2-4195ccc27e79 · outbound

This paper cites Generalized Preference Optimization: A Unified Approach to Offline Alignment.

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability Generalized Preference Optimization: A Unified Approach to Offline Alignment

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-12T05:43:32.031242Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4ae242d4-d980-40be-bae5-620781c02fea · outbound

This paper cites Improving Factuality in Large Language Models via Decoding-Time Hallucinatory and Truthful Comparators.

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability Improving Factuality in Large Language Models via Decoding-Time Hallucinatory and Truthful Comparators

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-12T05:43:32.050506Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8c79e017-8a33-482f-afa1-e6457421ce9a · outbound

This paper cites Tlcr: Token-level continuous reward for fine-grained reinforcement learning from human feedback.

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability Tlcr: Token-level continuous reward for fine-grained reinforcement learning from human feedback

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T05:43:33.403865Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 680747cf-b625-4563-b078-da4f0e69cde0 · outbound

This paper cites Scaling Relationship on Learning Mathematical Reasoning with Large Language Models.

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability Scaling Relationship on Learning Mathematical Reasoning with Large Language Models

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-12T05:43:32.074920Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 412575e6-f16c-4f16-b4a6-be6655f7d600 · outbound

This paper cites Token-level Direct Preference Optimization.

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability Token-level Direct Preference Optimization

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-12T05:43:32.080953Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation faf8ac9b-9b10-4ae9-b449-f34530718173 · outbound

This paper cites Alleviating Hallucinations of Large Language Models through Induced Hallucinations.

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability Alleviating Hallucinations of Large Language Models through Induced Hallucinations

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-12T05:43:32.128206Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0099b755-b59f-4b45-ae1c-3c8d4dfa0c37 · outbound

This paper cites Adversarial contrastive decoding: Boosting safety alignment of large language models via opposite prompt optimization.

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability Adversarial contrastive decoding: Boosting safety alignment of large language models via opposite prompt optimization

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-12T05:43:32.232648Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation afcbec82-3ff5-42d2-bbca-24375729ce36 · outbound

This paper cites Fine-Tuning Language Models from Human Preferences.

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability Fine-Tuning Language Models from Human Preferences

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-12T05:43:32.296269Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8bfd76e0-27fd-4260-830f-4685f2f4eb0b · outbound

This paper cites Training Verifiers to Solve Math Word Problems.

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability Training Verifiers to Solve Math Word Problems

Reference 2017

Resolution: unresolved; no resolver link, observed 2026-08-12T05:43:31.318039Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1c7f1a85-a33b-4fb7-b807-b808c40c1571 · outbound

This paper cites Decoding by Contrasting Knowledge: Enhancing LLMs' Confidence on Edited Facts.

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability Decoding by Contrasting Knowledge: Enhancing LLMs' Confidence on Edited Facts

Reference 2022

Resolution: unresolved; no resolver link, observed 2026-08-12T05:43:31.306581Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8b06d89d-3f53-4271-96e6-85480b26305a · outbound

This paper cites The Llama 3 Herd of Models.

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability The Llama 3 Herd of Models

Reference 2023

Resolution: unresolved; no resolver link, observed 2026-08-12T05:43:31.323256Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e7e60f8e-66ab-458b-be00-c1b5ad2aabe8 · outbound

This paper cites Direct Preference Optimization with an Offset.

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability Direct Preference Optimization with an Offset

Reference 2024

Resolution: unresolved; no resolver link, observed 2026-08-12T05:43:31.204558Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation 294c4aeb-1d41-4e1a-8897-da44984656a7 · inbound

Estimating LLM Uncertainty with Evidence cites this paper.

Estimating LLM Uncertainty with Evidence Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-09T19:37:15.914764Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4dc331b4-4341-4494-8e12-08eb8254157a · inbound

Probability-Consistent Preference Optimization for Enhanced LLM Reasoning cites this paper.

Probability-Consistent Preference Optimization for Enhanced LLM Reasoning Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-07T12:48:50.846271Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 570faee5-66db-4c4e-812c-0e3a9d8f6b1c · inbound

Reinforcing Video Reasoning with Focused Thinking cites this paper.

Reinforcing Video Reasoning with Focused Thinking Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-07T12:22:12.794644Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8892f5b0-4655-4877-9ffb-5ba0cfa528ae · inbound

Beyond the 80/20 Rule: High-Entropy Minority Tokens Drive Effective Reinforcement Learning for LLM Reasoning cites this paper.

Beyond the 80/20 Rule: High-Entropy Minority Tokens Drive Effective Reinforcement Learning for LLM Reasoning Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability

Reference 14

Resolution: verified exact; arxiv_id, observed 2026-05-12T12:12:08.896769Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation be8cfd95-abeb-4958-a511-bf33a06779c9 · inbound

Demystifying Reasoning Dynamics with Mutual Information: Thinking Tokens are Information Peaks in LLM Reasoning cites this paper.

Demystifying Reasoning Dynamics with Mutual Information: Thinking Tokens are Information Peaks in LLM Reasoning Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-07T11:19:46.616753Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 926f0a81-4344-4a05-a07c-9d898abd7893 · inbound

RAST: Reasoning Activation in LLMs via Small-model Transfer cites this paper.

RAST: Reasoning Activation in LLMs via Small-model Transfer Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability

Reference 34

Resolution: unresolved; no resolver link, observed 2026-08-07T12:19:55.541603Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 51e30123-4810-4801-b43d-cf1aaf244086 · inbound

Attention Illuminates LLM Reasoning: The Preplan-and-Anchor Rhythm Enables Fine-Grained Policy Optimization cites this paper.

Attention Illuminates LLM Reasoning: The Preplan-and-Anchor Rhythm Enables Fine-Grained Policy Optimization Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-04T09:48:06.827395Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c5c76e98-2b08-4706-9725-3efa6118ceb1 · inbound

Embarrassingly Simple Self-Distillation Improves Code Generation cites this paper.

Embarrassingly Simple Self-Distillation Improves Code Generation Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability

Reference 29

Resolution: unresolved; no resolver link, observed 2026-07-13T14:33:35.834383Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 60ebc9c1-3b5b-4a3f-9af1-27f1c10791db · inbound

AtManRL: Towards Faithful Reasoning via Differentiable Attention Saliency cites this paper.

AtManRL: Towards Faithful Reasoning via Differentiable Attention Saliency Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability

Reference 12

Resolution: verified exact; arxiv_id, observed 2026-05-10T08:22:37.719142Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation c373ab1f-1fda-41f9-993f-7c6d710cb6e6 · inbound

Select to Think: Unlocking SLM Potential with Local Sufficiency cites this paper.

Select to Think: Unlocking SLM Potential with Local Sufficiency Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability

Reference 11

Resolution: verified exact; arxiv_id, observed 2026-05-12T09:56:27.629689Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 4275dba6-da6f-4463-aef4-5c86ef9df97a · inbound

Select to Think: Unlocking SLM Potential with Local Sufficiency cites this paper.

Select to Think: Unlocking SLM Potential with Local Sufficiency Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability

Reference 11

Resolution: verified exact; arxiv_id, observed 2026-07-01T08:35:33.657546Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 129a9b1a-b312-4ae1-bb0f-4b6908089bc8 · inbound

When Are Experts Misrouted? Counterfactual Routing Analysis in Mixture-of-Experts Language Models cites this paper.

When Are Experts Misrouted? Counterfactual Routing Analysis in Mixture-of-Experts Language Models Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability

Reference 38

Resolution: verified exact; arxiv_id, observed 2026-05-11T01:50:51.040367Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 789b8bda-373b-47a3-8eb4-797ec833a81b · inbound

Your Language Model is Its Own Critic: Reinforcement Learning with Value Estimation from Actor's Internal States cites this paper.

Your Language Model is Its Own Critic: Reinforcement Learning with Value Estimation from Actor's Internal States Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability

Reference 17

Resolution: metadata mismatch; arxiv_id, observed 2026-05-11T03:55:54.847622Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation bea7e6b6-1107-406c-85fd-1e971aa23b40 · inbound

Your Language Model is Its Own Critic: Reinforcement Learning with Value Estimation from Actor's Internal States cites this paper.

Reference 17

Resolution: metadata mismatch; arxiv_id, observed 2026-05-12T06:41:45.912311Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 6acf0d4d-e4e7-4a93-bbe8-04ae7f564e4d · inbound

HTPO: Towards Exploration-Exploitation Balanced Policy Optimization via Hierarchical Token-level Objective Control cites this paper.

HTPO: Towards Exploration-Exploitation Balanced Policy Optimization via Hierarchical Token-level Objective Control Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability

Reference 14

Resolution: verified exact; arxiv_id, observed 2026-05-12T00:51:14.635110Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 29d311ec-69c6-4df4-8f62-a0ad9db35357 · inbound

Stateful Reasoning via Insight Replay cites this paper.

Stateful Reasoning via Insight Replay Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability

Reference 17

Resolution: verified exact; arxiv_id, observed 2026-05-15T01:43:27.643827Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 1611c923-e0d4-48d7-a207-bd1c408213cc · inbound

Stateful Reasoning via Insight Replay cites this paper.

Stateful Reasoning via Insight Replay Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability

Reference 17

Resolution: verified exact; arxiv_id, observed 2026-05-20T21:39:03.356197Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 45fff654-df73-47ca-822f-caae21c4d517 · inbound

Token-weighted Direct Preference Optimization with Attention cites this paper.

Token-weighted Direct Preference Optimization with Attention Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability

Reference 51

Resolution: verified exact; arxiv_id, observed 2026-05-22T07:06:11.755054Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 8f90ebb5-37ac-4ce0-8576-473783e46ae8 · inbound

Token-weighted Direct Preference Optimization with Attention cites this paper.

Token-weighted Direct Preference Optimization with Attention Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability

Reference 1

Resolution: metadata mismatch; arxiv_id, observed 2026-07-01T15:05:48.155143Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation a4016e76-460e-4422-a3b5-7515261b4ea6 · inbound

Smaller Models are Natural Explorers for Policy-Level Diversity in GRPO cites this paper.

Smaller Models are Natural Explorers for Policy-Level Diversity in GRPO Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability

Reference 21

Resolution: verified exact; arxiv_id, observed 2026-06-29T00:02:49.986417Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 4eb04468-593f-41fe-84a8-8aa6c8dd8d53 · inbound

Thinking Economically: A Hierarchical Framework for Adaptive-Complexity Reasoning in LLMs cites this paper.

Thinking Economically: A Hierarchical Framework for Adaptive-Complexity Reasoning in LLMs Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability

Reference 45

Resolution: verified exact; arxiv_id, observed 2026-06-28T17:12:25.275722Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 78698ae4-996a-4c41-a85e-dd2adc39d311 · inbound

The Tell-Tale Norm: $\ell_2$ Magnitude as a Signal for Reasoning Dynamics in Large Language Models cites this paper.

The Tell-Tale Norm: $\ell_2$ Magnitude as a Signal for Reasoning Dynamics in Large Language Models Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability

Reference 2

Resolution: verified exact; arxiv_id, observed 2026-07-02T12:26:56.922063Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation b56ec185-33d7-4752-8fb3-4665efe65114 · inbound

Sample Where You Struggle: Sharpening Base Model Reasoning via Entropy-Guided Power Sampling cites this paper.

Sample Where You Struggle: Sharpening Base Model Reasoning via Entropy-Guided Power Sampling Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability

Reference 12

Resolution: verified exact; arxiv_id, observed 2026-07-02T22:37:25.538625Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 581eb95a-5873-42cc-b4ab-c9132a3eee2e · inbound

The Periodic Table of LLM Reasoning: A Structured Survey of Reasoning Paradigms, Methods, and Failure Modes cites this paper.

The Periodic Table of LLM Reasoning: A Structured Survey of Reasoning Paradigms, Methods, and Failure Modes Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability

Reference 149

Resolution: metadata mismatch; arxiv_id, observed 2026-07-03T05:57:41.329712Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 48859137-e761-451f-8a44-ae33c494db81 · inbound

Beyond Fully Random Masking: Attention-Guided Denoising and Optimization for Diffusion Language Models cites this paper.

Beyond Fully Random Masking: Attention-Guided Denoising and Optimization for Diffusion Language Models Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability

Reference 53

Resolution: verified exact; arxiv_id, observed 2026-07-03T11:28:04.395213Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 7831af7e-d003-4267-890d-3cfe74406dc9 · inbound

Beyond Entropy: Learning from Token-Level Distributional Deviations for LLM Reasoning cites this paper.

Beyond Entropy: Learning from Token-Level Distributional Deviations for LLM Reasoning Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability

Reference 11

Resolution: metadata mismatch; arxiv_id, observed 2026-07-04T03:49:30.155417Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation bdadb457-eb31-4c85-8b4e-19aedac412ad · inbound

Rethinking On-Policy Self-Distillation for Thinking Models cites this paper.

Rethinking On-Policy Self-Distillation for Thinking Models Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability

Reference 7

Resolution: verified exact; local_arxiv, observed 2026-07-08T01:14:27.539267Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.