·
DataBubble
  • Home
  • Models
  • News
  • Compare
  • Boards
  • Pricing
  • About
  • Newsletter
  • Methodology
  • Contact
Latest
CAPER: Clause-Aligned Process Supervision for Text-to-SQL1h◆Skill-RM: Unifying Heterogeneous Evaluation Criteria via Agent Skill1h◆Neuron Populations Exhibit Divergent Selectivity with Scale1h◆Evaluating the Reversal Curse in Model Editing1h◆Ranking Free RAG: Replacing Re-ranking with Selection in RAG for Sensitive Domains1h◆Enhancing Paraphrase Type Generation: The Impact of DPO and RLHF Evaluated with Human-Ranked Data1h◆Learning without training: The implicit dynamics of in-context learning1h◆Privacy-Aware Decoding: Mitigating Privacy Leakage of Large Language Models in Retrieval-Augmented Generation1h◆CTR-Sink: Attention Sink for Language Models in Click-Through Rate Prediction1h◆When Models Refuse: Political Steerability and Feature Richness as Measures of Ideological Depth1h◆Train Once, Reuse Everywhere: Generalizable Implicit In-Context Learning by Routing Attention1h◆Instant Personalized Large Language Model Adaptation via Hypernetwork1h◆REFLEX: Self-Refining Explainable Fact-Checking via Verdict-Anchored Style Control1h◆CourseTimeQA: A Lecture-Video Benchmark and a Latency-Constrained Cross-Modal Fusion Method for Timestamped QA1h◆KBQA-R1: Reinforcing Large Language Models for Knowledge Base Question Answering1h◆Social Caption: Evaluating Social Understanding in Multimodal Models1h◆Neural Attention Search Linear: Towards Adaptive Token-Level Hybrid Attention Models1h◆Embracing Anisotropy: Turning Massive Activations into Interpretable Control Knobs for Large Language Models1h◆Can Structural Cues Save LLMs? Evaluating Language Models in Massive Document Streams1h◆Learning Concepts, Not Tokens: Self-Supervised Semantic Alignment for Language Models1h◆CAPER: Clause-Aligned Process Supervision for Text-to-SQL1h◆Skill-RM: Unifying Heterogeneous Evaluation Criteria via Agent Skill1h◆Neuron Populations Exhibit Divergent Selectivity with Scale1h◆Evaluating the Reversal Curse in Model Editing1h◆Ranking Free RAG: Replacing Re-ranking with Selection in RAG for Sensitive Domains1h◆Enhancing Paraphrase Type Generation: The Impact of DPO and RLHF Evaluated with Human-Ranked Data1h◆Learning without training: The implicit dynamics of in-context learning1h◆Privacy-Aware Decoding: Mitigating Privacy Leakage of Large Language Models in Retrieval-Augmented Generation1h◆CTR-Sink: Attention Sink for Language Models in Click-Through Rate Prediction1h◆When Models Refuse: Political Steerability and Feature Richness as Measures of Ideological Depth1h◆Train Once, Reuse Everywhere: Generalizable Implicit In-Context Learning by Routing Attention1h◆Instant Personalized Large Language Model Adaptation via Hypernetwork1h◆REFLEX: Self-Refining Explainable Fact-Checking via Verdict-Anchored Style Control1h◆CourseTimeQA: A Lecture-Video Benchmark and a Latency-Constrained Cross-Modal Fusion Method for Timestamped QA1h◆KBQA-R1: Reinforcing Large Language Models for Knowledge Base Question Answering1h◆Social Caption: Evaluating Social Understanding in Multimodal Models1h◆Neural Attention Search Linear: Towards Adaptive Token-Level Hybrid Attention Models1h◆Embracing Anisotropy: Turning Massive Activations into Interpretable Control Knobs for Large Language Models1h◆Can Structural Cues Save LLMs? Evaluating Language Models in Massive Document Streams1h◆Learning Concepts, Not Tokens: Self-Supervised Semantic Alignment for Language Models1h◆
News/Configurable Reward Model for Balanced Safety Alignment
arxiv
PublishedJune 1, 2026 at 4:00 AM

Configurable Reward Model for Balanced Safety Alignment

Source
arxiv.orgfull article ↗
Read on arxiv→
Publisher summary· verbatim

arXiv:2605.30487v1 Announce Type: new Abstract: Aligning large language models (LLMs) to heterogeneous and rapidly evolving safety requirements remains a critical challenge. Existing instruction-tuned LLMs and standalone safety classifiers often fail to generalize to new safety configurations, motiv

Stay posted· Newsletter

A 5-min weekly brief — top movers, price watch, story of the week.

// no spam · unsubscribe one-click · free forever

Discussion
Source
↗
arxiv
Read original ↗All from arxiv →

No replies yet. Be first.

Source
↗
arxiv
Read original ↗All from arxiv →

Related coverage

More from ARXIV
arxivCAPER: Clause-Aligned Process Supervision for Text-to-SQL1harxivSkill-RM: Unifying Heterogeneous Evaluation Criteria via Agent Skill1harxivNeuron Populations Exhibit Divergent Selectivity with Scale1harxivEvaluating the Reversal Curse in Model Editing1h
The Bubble Brief
WEEKLY

Read AI insights every Tuesday — top movers, new releases, story of the week.

// no spam · unsubscribe one-click · free forever

Originally published on arxiv ↗
HomeModelsNews