UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma | Paper Citation Record

Observation 6ecaa784-1b7b-45af-a597-0c1b5a8609a2 · outbound

This paper cites Back to basics: Revisiting reinforce-style optimization for learning from human feedback in llms.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Back to basics: Revisiting reinforce-style optimization for learning from human feedback in llms

Reference 1

Resolution: verified fuzzy; raw_fallback, observed 2026-07-09T22:26:37.546147Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 2d1d6d8d-9ef5-4661-8a4b-9327ceb38a45 · outbound

This paper cites Constitutional AI: Harmlessness from AI Feedback.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Constitutional AI: Harmlessness from AI Feedback

Reference 2

Resolution: verified exact; local_arxiv, observed 2026-07-09T22:26:37.194238Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 97768206-e2e7-458a-9f90-51011be5ec09 · outbound

This paper cites Efficient reinforcement learning with semantic and token entropy for llm reasoning.arXiv preprint arXiv:2512.04359, 2025.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Efficient reinforcement learning with semantic and token entropy for llm reasoning.arXiv preprint arXiv:2512.04359, 2025

Reference 3

Resolution: verified exact; arxiv_id, observed 2026-07-09T22:26:37.144898Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 43d3cb3c-2cbf-48be-a23e-6e0625c73da5 · outbound

This paper cites MiniMax-M1: Scaling Test-Time Compute Efficiently with Lightning Attention.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma MiniMax-M1: Scaling Test-Time Compute Efficiently with Lightning Attention

Reference 4

Resolution: verified exact; local_arxiv, observed 2026-07-09T22:26:37.205825Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 6b1f59e4-5202-4fe7-986e-3fd57d6a8090 · outbound

This paper cites American invitational mathematics examination-aime 2024.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma American invitational mathematics examination-aime 2024

Reference 5

Resolution: verified fuzzy; raw_fallback, observed 2026-07-09T22:26:37.547848Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 84f76c6a-8d46-4b69-a342-163eef14c5b3 · outbound

This paper cites Rethinking Muon Beyond Pretraining: Spectral Failures and High-Pass Remedies for VLA and RLVR.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Rethinking Muon Beyond Pretraining: Spectral Failures and High-Pass Remedies for VLA and RLVR

Reference 6

Resolution: verified exact; local_arxiv, observed 2026-07-09T22:26:37.172379Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 9c6e3eb3-76f2-41ff-92a6-64d5813f8103 · outbound

This paper cites Hero, and Sijia Liu.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Hero, and Sijia Liu

Reference 7

Resolution: verified fuzzy; raw_fallback, observed 2026-07-09T22:26:37.558585Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 8cd12236-c4cb-407c-8299-b389a8f2607f · outbound

This paper cites an unresolved cited work.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Unresolved cited work

Reference 8

Resolution: unresolved; raw_fallback, observed 2026-07-09T22:26:37.570176Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 05abc2b9-07d0-4070-a40b-24c2660f3ef5 · outbound

This paper cites Soft Adaptive Policy Optimization.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Soft Adaptive Policy Optimization

Reference 9

Resolution: verified exact; local_arxiv, observed 2026-07-09T22:26:37.220833Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation e2e5cef2-b1a7-4c72-b972-d78c584c3e88 · outbound

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

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning

Reference 10

Resolution: verified exact; local_arxiv, observed 2026-07-09T22:26:37.178406Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 188c66a1-c992-44cb-ba81-5a4763573b8b · outbound

This paper cites Olympiadbench: A challenging benchmark for promoting agi with olympiad-level bilingual multimodal scientific problems.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Olympiadbench: A challenging benchmark for promoting agi with olympiad-level bilingual multimodal scientific problems

Reference 11

Resolution: verified fuzzy; raw_fallback, observed 2026-07-09T22:26:37.556925Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation a638b56e-b728-4e8a-beb6-e719adf54287 · outbound

This paper cites REINFORCE++: Stabilizing Critic-Free Policy Optimization with Global Advantage Normalization.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma REINFORCE++: Stabilizing Critic-Free Policy Optimization with Global Advantage Normalization

Reference 12

Resolution: verified exact; local_arxiv, observed 2026-07-09T22:26:37.184446Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation ed73da8b-ba1d-4468-bba3-9be437561718 · outbound

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

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Efficient memory management for large language model serving with pagedattention

Reference 13

Resolution: verified fuzzy; raw_fallback, observed 2026-07-09T22:26:37.560237Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 46b2b7b2-b5f5-41c7-96ad-62e48316a218 · outbound

This paper cites Solving quantitative reasoning problems with language models.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Solving quantitative reasoning problems with language models

Reference 14

Resolution: verified fuzzy; raw_fallback, observed 2026-07-09T22:26:37.551171Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 8210591f-20ee-4bc1-b912-975aaa72e5e7 · outbound

This paper cites Back to basics: Revisiting exploration in reinforcement learning for llm reasoning via generative probabilities.arXiv preprint arXiv:2602.05281.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Back to basics: Revisiting exploration in reinforcement learning for llm reasoning via generative probabilities.arXiv preprint arXiv:2602.05281

Reference 15

Resolution: verified exact; arxiv_id, observed 2026-07-09T22:26:37.157612Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 1af9b955-f292-4b10-9334-8e17eeb058ec · outbound

This paper cites Bandpo: Bridging trust regions and ratio clipping via probability-aware bounds for llm reinforcement learning.arXiv preprint arXiv:2603.04918, 2026.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Bandpo: Bridging trust regions and ratio clipping via probability-aware bounds for llm reinforcement learning.arXiv preprint arXiv:2603.04918, 2026

Reference 16

Resolution: verified exact; arxiv_id, observed 2026-07-09T22:26:37.163679Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation c424ffdf-b5d7-4941-8fbf-d34faa72f81e · outbound

This paper cites Let’s verify step by step.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Let’s verify step by step

Reference 17

Resolution: verified fuzzy; raw_fallback, observed 2026-07-09T22:26:37.561765Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation a73a5e02-b75f-4aa2-9b0f-97f26eabc1bc · outbound

This paper cites Length-unbiased sequence policy optimization: Revealing and controlling response length variation in rlvr.arXiv preprint arXiv:2602.05261.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Length-unbiased sequence policy optimization: Revealing and controlling response length variation in rlvr.arXiv preprint arXiv:2602.05261

Reference 18

Resolution: verified exact; arxiv_id, observed 2026-07-09T22:26:37.208422Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 1e606b2d-4d75-4a88-9c61-66afb52ae5f8 · outbound

This paper cites Understanding R1-Zero-Like Training: A Critical Perspective.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Understanding R1-Zero-Like Training: A Critical Perspective

Reference 19

Resolution: verified exact; local_arxiv, observed 2026-07-09T22:26:37.169302Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation dad217d4-b78d-4c7e-9ad1-4b06b0080d45 · outbound

This paper cites Inter-GPS: Interpretable geometry problem solving with formal language and symbolic reasoning.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Inter-GPS: Interpretable geometry problem solving with formal language and symbolic reasoning

Reference 20

Resolution: verified fuzzy; raw_fallback, observed 2026-07-09T22:26:37.549493Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 307656d7-c748-419f-9431-0a0caaa19fe3 · outbound

This paper cites Stabilizing MoE Reinforcement Learning by Aligning Training and Inference Routers.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Stabilizing MoE Reinforcement Learning by Aligning Training and Inference Routers

Reference 21

Resolution: verified exact; arxiv_id, observed 2026-07-09T22:26:37.154897Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 5752603f-8b67-48e7-b95e-6c9657a8108a · outbound

This paper cites Clip your sequences fairly: Enforcing length fairness for sequence-level rl.arXiv preprint arXiv:2509.09177.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Clip your sequences fairly: Enforcing length fairness for sequence-level rl.arXiv preprint arXiv:2509.09177

Reference 22

Resolution: verified exact; arxiv_id, observed 2026-07-09T22:26:37.152364Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 85a27c07-f2b0-4cb9-adb9-bb2e7b5b1be5 · outbound

This paper cites Training language models to follow instructions with human feedback.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Training language models to follow instructions with human feedback

Reference 23

Resolution: verified fuzzy; raw_fallback, observed 2026-07-09T22:26:37.544379Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 6c9d7f76-ac73-4975-b320-deeb51bd696d · outbound

This paper cites Rethinking the Trust Region in LLM Reinforcement Learning.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Rethinking the Trust Region in LLM Reinforcement Learning

Reference 24

Resolution: verified exact; local_arxiv, observed 2026-07-09T22:26:37.210776Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation f9bf1274-34d2-4bcd-b657-6a9ea585f631 · outbound

This paper cites Tapered Off-Policy REINFORCE: Stable and efficient reinforcement learning for LLMs.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Tapered Off-Policy REINFORCE: Stable and efficient reinforcement learning for LLMs

Reference 25

Resolution: verified exact; local_arxiv, observed 2026-07-09T22:26:37.213079Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 6d30ce88-df57-4bcd-b92d-b05c0d44fdb4 · outbound

This paper cites Trustregionpolicyoptimization.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Trustregionpolicyoptimization

Reference 26

Resolution: verified fuzzy; raw_fallback, observed 2026-07-09T22:26:37.566613Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 81b169db-2a72-4d47-9a77-fcf3382ee7a8 · outbound

This paper cites Proximal Policy Optimization Algorithms.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Proximal Policy Optimization Algorithms

Reference 27

Resolution: verified exact; local_arxiv, observed 2026-07-09T22:26:37.147239Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 4e27c231-70d3-4c1d-94d4-086dce0ae3ab · outbound

This paper cites Seed2.0 Model Card: Towards Intelligence Frontier for Real-World Complexity.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Seed2.0 Model Card: Towards Intelligence Frontier for Real-World Complexity

Reference 28

Resolution: verified exact; local_arxiv, observed 2026-07-09T22:26:37.203495Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation b0ca443f-67e5-4285-b13c-b834c9c04eb1 · outbound

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

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models

Reference 29

Resolution: verified exact; local_arxiv, observed 2026-07-09T22:26:37.149556Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation f24699ae-8f49-4063-a8fd-548d4f44039d · outbound

This paper cites Hybridflow: A flexible and efficient rlhf framework.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Hybridflow: A flexible and efficient rlhf framework

Reference 30

Resolution: verified fuzzy; raw_fallback, observed 2026-07-09T22:26:37.563365Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation df14d639-5808-4c95-ab37-e8443cd1c7ff · outbound

This paper cites Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters

Reference 31

Resolution: verified exact; local_arxiv, observed 2026-07-09T22:26:37.181003Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 3c35c6cd-3213-478b-acbb-033b90cb0194 · outbound

This paper cites Klear-reasoner: Advancing reasoning capability via gradient-preserving clipping policy optimization.arXiv preprint arXiv:2508.07629.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Klear-reasoner: Advancing reasoning capability via gradient-preserving clipping policy optimization.arXiv preprint arXiv:2508.07629

Reference 32

Resolution: verified exact; arxiv_id, observed 2026-07-09T22:26:37.166428Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation f62037ce-208b-412f-8689-8761caf75da3 · outbound

This paper cites When Importance Sampling Misallocates Credit: Asymmetric Ratios for Outcome-Supervised RL.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma When Importance Sampling Misallocates Credit: Asymmetric Ratios for Outcome-Supervised RL

Reference 33

Resolution: verified exact; local_arxiv, observed 2026-07-09T22:26:37.218159Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 30250df8-2189-400b-b35a-791e85d7d501 · outbound

This paper cites Simple statistical gradient-following algorithms for connectionist reinforcement learning.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Simple statistical gradient-following algorithms for connectionist reinforcement learning

Reference 34

Resolution: verified fuzzy; raw_fallback, observed 2026-07-09T22:26:37.568414Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 11e7ffd4-62da-4fb0-ad77-3e0b2b1ce165 · outbound

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

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Qwen2.5-Math Technical Report: Toward Mathematical Expert Model via Self-Improvement

Reference 35

Resolution: verified exact; local_arxiv, observed 2026-07-09T22:26:37.189407Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 5c3659a4-0f46-48e2-8f28-4d8e9753cdd4 · outbound

This paper cites Qwen3 Technical Report.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Qwen3 Technical Report

Reference 36

Resolution: verified exact; local_arxiv, observed 2026-07-09T22:26:37.196554Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 7baf2c4e-27fe-4c60-bb29-22add431be8f · outbound

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

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma DAPO: An Open-Source LLM Reinforcement Learning System at Scale

Reference 37

Resolution: verified exact; local_arxiv, observed 2026-07-09T22:26:37.198934Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 118ea934-f2e3-4e36-90a5-a9c9bb703074 · outbound

This paper cites Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?

Reference 38

Resolution: verified exact; local_arxiv, observed 2026-07-09T22:26:37.223302Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 802ea673-266f-46f2-80bd-7c5365a9ea4d · outbound

This paper cites Star: Bootstrapping reasoning with reasoning.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Star: Bootstrapping reasoning with reasoning

Reference 39

Resolution: verified fuzzy; raw_fallback, observed 2026-07-09T22:26:37.565043Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation b6faac81-0a95-46fa-b6b0-1ed56c35f881 · outbound

This paper cites George E Uhlenbeck and Leonard S Ornstein.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma George E Uhlenbeck and Leonard S Ornstein

Reference 40

Resolution: metadata mismatch; arxiv_id, observed 2026-07-09T22:26:37.187015Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 73096a43-0a93-48c7-b6ab-9b570dd25d25 · outbound

This paper cites Geometric-mean policy optimization.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Geometric-mean policy optimization

Reference 41

Resolution: verified exact; arxiv_id, observed 2026-07-09T22:26:37.192014Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 31dc8757-8ed5-4869-9478-1e6dd1adb69c · outbound

This paper cites Group Sequence Policy Optimization.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Group Sequence Policy Optimization

Reference 42

Resolution: verified exact; local_arxiv, observed 2026-07-09T22:26:37.201213Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 70b4d23e-3657-4bc5-bef7-a357d4b22626 · outbound

This paper cites Prosperity before collapse: How far can off-policy rl reach with stale data on llms?.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma Prosperity before collapse: How far can off-policy rl reach with stale data on llms?

Reference 43

Resolution: verified exact; arxiv_id, observed 2026-07-09T22:26:37.215591Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 22670313-d73a-4f44-9de9-f7a4f3f27c49 · outbound

This paper cites The surprising effectiveness of negative reinforcement in llm reasoning.arXiv preprint arXiv:2506.01347.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma The surprising effectiveness of negative reinforcement in llm reasoning.arXiv preprint arXiv:2506.01347

Reference 44

Resolution: verified exact; arxiv_id, observed 2026-07-09T22:26:37.175791Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation ea02100c-396a-458b-8896-cb07df26438e · outbound

This paper cites If πθ < π lower, the gradient is nullified to prevent the probability from dropping too low (over-penalization).

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma If πθ < π lower, the gradient is nullified to prevent the probability from dropping too low (over-penalization)

Reference 45

Resolution: verified fuzzy; raw_fallback, observed 2026-07-09T22:26:37.555237Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation c478bb1e-f250-47ed-a885-f0c332ddbdec · outbound

This paper cites 1 G GX i=1 ˆAi 1 sg(πθ(oi|q)) 1 |oi | ∇θ πθ(oi|q) 1 |oi | # =E q,{oi}.

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma 1 G GX i=1 ˆAi 1 sg(πθ(oi|q)) 1 |oi | ∇θ πθ(oi|q) 1 |oi | # =E q,{oi}

Reference 46

Resolution: verified fuzzy; raw_fallback, observed 2026-07-09T22:26:37.553040Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Coverage vector

Reference resolution

Outbound references

Pith citing papers