Cover Image

PRISM: Probability Reallocation with In-Span Masking for Knowledge-Sensitive Alignment

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Xu, Chenning, Zheng, Mao, Song, Mingyang
Format:	Preprint
Published:	2026
Subjects:	Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Beyond the Illusion of Consensus: From Surface Heuristics to Knowledge-Grounded Evaluation in LLM-as-a-Judge
by: Song, Mingyang, et al.
Published: (2026)

PodBench: A Comprehensive Benchmark for Instruction-Aware Audio-Oriented Podcast Script Generation
by: Xu, Chenning, et al.
Published: (2026)

TAT-R1: Terminology-Aware Translation with Reinforcement Learning and Word Alignment
by: Li, Zheng, et al.
Published: (2025)

HardMTBench: Stress-Testing Chinese-English Translation on Knowledge-Intensive Domains
by: Li, Zheng, et al.
Published: (2026)

Model Merging in the Era of Large Language Models: Methods, Applications, and Future Directions
by: Song, Mingyang, et al.
Published: (2026)

Walk Before You Run! Concise LLM Reasoning via Reinforcement Learning
by: Song, Mingyang, et al.
Published: (2025)

A Survey of Query Optimization in Large Language Models
by: Song, Mingyang, et al.
Published: (2024)

A Survey of On-Policy Distillation for Large Language Models
by: Song, Mingyang, et al.
Published: (2026)

Counting-Stars: A Multi-evidence, Position-aware, and Scalable Benchmark for Evaluating Long-Context Large Language Models
by: Song, Mingyang, et al.
Published: (2024)

GRP: Goal-Reversed Prompting for Zero-Shot Evaluation with LLMs
by: Song, Mingyang, et al.
Published: (2025)

MiMoTable: A Multi-scale Spreadsheet Benchmark with Meta Operations for Table Reasoning
by: Li, Zheng, et al.
Published: (2024)

Can Many-Shot In-Context Learning Help LLMs as Evaluators? A Preliminary Empirical Study
by: Song, Mingyang, et al.
Published: (2024)

HY-MT1.5 Technical Report
by: Zheng, Mao, et al.
Published: (2025)

CodeDelegator: Mitigating Context Pollution via Role Separation in Code-as-Action Agents
by: Fei, Tianxiang, et al.
Published: (2026)

CONTESTS: a Framework for Consistency Testing of Span Probabilities in Language Models
by: Wagner, Eitan, et al.
Published: (2024)

SRA: Span Representation Alignment for Large Language Model Distillation
by: Dao, Quoc Phong, et al.
Published: (2026)

The PRISM Alignment Dataset: What Participatory, Representative and Individualised Human Feedback Reveals About the Subjective and Multicultural Alignment of Large Language Models
by: Kirk, Hannah Rose, et al.
Published: (2024)

DisGeM: Distractor Generation for Multiple Choice Questions with Span Masking
by: Cavusoglu, Devrim, et al.
Published: (2024)

SAG: Style-Aligned Article Generation via Model Collaboration
by: Xu, Chenning, et al.
Published: (2024)

SSR-Zero: Simple Self-Rewarding Reinforcement Learning for Machine Translation
by: Yang, Wenjie, et al.
Published: (2025)

Hunyuan-MT Technical Report
by: Zheng, Mao, et al.
Published: (2025)

FastCuRL: Curriculum Reinforcement Learning with Stage-wise Context Scaling for Efficient Training R1-like Reasoning Models
by: Song, Mingyang, et al.
Published: (2025)

IFMTBench: A Comprehensive Benchmark for Multilingual Translation Instruction Following
by: Sun, Mingrui, et al.
Published: (2026)

Temporal Alignment of Time Sensitive Facts with Activation Engineering
by: Govindan, Sanjay, et al.
Published: (2025)

Half the Nonlinearity Is Wasted: Measuring and Reallocating the Transformer's MLP Budget
by: Balogh, Peter
Published: (2026)

PRISM of Opinions: A Persona-Reasoned Multimodal Framework for User-centric Conversational Stance Detection
by: Wang, Bingbing, et al.
Published: (2025)

On the Entity-Level Alignment in Crosslingual Consistency
by: Liu, Yihong, et al.
Published: (2025)

Contrastive Knowledge Transfer and Robust Optimization for Secure Alignment of Large Language Models
by: Zheng, Jiasen, et al.
Published: (2025)

PRISM-MCTS: Learning from Reasoning Trajectories with Metacognitive Reflection
by: Cheng, Siyuan, et al.
Published: (2026)

MockConf: A Student Interpretation Dataset: Analysis, Word- and Span-level Alignment and Baselines
by: Javorský, Dávid, et al.
Published: (2025)

PRISM: Parametrically Refactoring Inference for Speculative Sampling Draft Models
by: Wang, Xuliang, et al.
Published: (2026)

Span Modeling for Idiomaticity and Figurative Language Detection with Span Contrastive Loss
by: Matheny, Blake, et al.
Published: (2026)

The Anatomy of an Edit: Mechanism-Guided Activation Steering for Knowledge Editing
by: Cao, Yuan, et al.
Published: (2026)

Multi-hop Reasoning via Early Knowledge Alignment
by: Wang, Yuxin, et al.
Published: (2025)

Mask-DPO: Generalizable Fine-grained Factuality Alignment of LLMs
by: Gu, Yuzhe, et al.
Published: (2025)

Assessing the Sensitivity and Alignment of FOL Closeness Metrics
by: Thatikonda, Ramya Keerthy, et al.
Published: (2025)

Training LLM-Based Agents with Synthetic Self-Reflected Trajectories and Partial Masking
by: Chen, Yihan, et al.
Published: (2025)

MisSpans: Fine-Grained False Span Identification in Cross-Domain Fake News
by: Liu, Zhiwei, et al.
Published: (2026)

A Probabilistic Choreography Language for PRISM
by: Carbone, Marco, et al.
Published: (2025)

Divide-Then-Align: Honest Alignment based on the Knowledge Boundary of RAG
by: Sun, Xin, et al.
Published: (2025)