Cover Image

Keep Calm and Avoid Harmful Content: Concept Alignment and Latent Manipulation Towards Safer Answers

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Belo, Ruben, Guimaraes, Marta, Soares, Claudia
Format:	Preprint
Published:	2025
Subjects:	Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs
by: Mendu, Sai Krishna, et al.
Published: (2025)

Harmful Visual Content Manipulation Matters in Misinformation Detection Under Multimedia Scenarios
by: Wang, Bing, et al.
Published: (2026)

Probability of Collision of satellites and space debris for short-term encounters: Rederivation and fast-to-compute upper and lower bounds
by: Ferreira, Ricardo, et al.
Published: (2023)

If Concept Bottlenecks are the Question, are Foundation Models the Answer?
by: Debole, Nicola, et al.
Published: (2025)

Deliberative Alignment: Reasoning Enables Safer Language Models
by: Guan, Melody Y., et al.
Published: (2024)

LookAhead Tuning: Safer Language Models via Partial Answer Previews
by: Liu, Kangwei, et al.
Published: (2025)

Towards Understanding and Avoiding Limitations of Convolutions on Graphs
by: Roth, Andreas
Published: (2026)

Generalizing Trilateration: Approximate Maximum Likelihood Estimator for Initial Orbit Determination in Low-Earth Orbit
by: Ferreira, Ricardo, et al.
Published: (2024)

Let it Calm: Exploratory Annealed Decoding for Verifiable Reinforcement Learning
by: Yang, Chenghao, et al.
Published: (2025)

Toward Safer Diffusion Language Models: Discovery and Mitigation of Priming Vulnerability
by: Yamabe, Shojiro, et al.
Published: (2025)

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning
by: Chen, Liang, et al.
Published: (2025)

Safety Representations for Safer Policy Learning
by: Mani, Kaustubh, et al.
Published: (2025)

Conformal Feedback Alignment: Quantifying Answer-Level Reliability for Robust LLM Alignment
by: Chen, Tiejin, et al.
Published: (2026)

Generalized Distributional Alignment Games for Unbiased Answer-Level Fine-Tuning
by: Mohri, Mehryar, et al.
Published: (2026)

Two Calm Ends and the Wild Middle: A Geometric Picture of Memorization in Diffusion Models
by: Dodson, Nick, et al.
Published: (2026)

SPAARS: Safer RL Policy Alignment through Abstract Exploration and Refined Exploitation of Action Space
by: K, Swaminathan S, et al.
Published: (2026)

Toward Real-World IoT Security: Concept Drift-Resilient IoT Botnet Detection via Latent Space Representation Learning and Alignment
by: Wasswa, Hassan, et al.
Published: (2025)

Uncertainty-aware Latent Safety Filters for Avoiding Out-of-Distribution Failures
by: Seo, Junwon, et al.
Published: (2025)

Adversarially Robust Detection of Harmful Online Content: A Computational Design Science Approach
by: Chai, Yidong, et al.
Published: (2025)

Model Stitching by Functional Latent Alignment
by: Athanasiadis, Ioannis, et al.
Published: (2025)

Concept Alignment
by: Rane, Sunayana, et al.
Published: (2024)

ChineseHarm-Bench: A Chinese Harmful Content Detection Benchmark
by: Liu, Kangwei, et al.
Published: (2025)

Nonparametric Identification of Latent Concepts
by: Zheng, Yujia, et al.
Published: (2025)

Consensus Sampling for Safer Generative AI
by: Kalai, Adam Tauman, et al.
Published: (2025)

Distributional Alignment Games for Answer-Level Fine-Tuning
by: Mohri, Mehryar, et al.
Published: (2026)

Lisa: Lazy Safety Alignment for Large Language Models against Harmful Fine-tuning Attack
by: Huang, Tiansheng, et al.
Published: (2024)

Latent Space Translation via Semantic Alignment
by: Maiorca, Valentino, et al.
Published: (2023)

Reasoned Safety Alignment: Ensuring Jailbreak Defense via Answer-Then-Check
by: Cao, Chentao, et al.
Published: (2025)

The Cost of Avoiding Backpropagation
by: Panchal, Kunjal, et al.
Published: (2025)

Avoiding Leakage Poisoning: Concept Interventions Under Distribution Shifts
by: Zarlenga, Mateo Espinosa, et al.
Published: (2025)

Exchangeable Sequence Models Quantify Uncertainty Over Latent Concepts
by: Ye, Naimeng, et al.
Published: (2024)

Toward Faithful and Complete Answer Construction from a Single Document
by: Chen, Zhaoyang, et al.
Published: (2026)

Latent Adversarial Training Improves Robustness to Persistent Harmful Behaviors in LLMs
by: Sheshadri, Abhay, et al.
Published: (2024)

Safer Policy Compliance with Dynamic Epistemic Fallback
by: Imperial, Joseph Marvin, et al.
Published: (2026)

Learning Discrete Concepts in Latent Hierarchical Models
by: Kong, Lingjing, et al.
Published: (2024)

Simple Mechanisms for Representing, Indexing and Manipulating Concepts
by: Li, Yuanzhi, et al.
Published: (2023)

Latent Adaptive Planner for Dynamic Manipulation
by: Noh, Donghun, et al.
Published: (2025)

SocialDF: Benchmark Dataset and Detection Model for Mitigating Harmful Deepfake Content on Social Media Platforms
by: Batra, Arnesh, et al.
Published: (2025)

Concept-Aware Latent and Explicit Knowledge Integration for Enhanced Cognitive Diagnosis
by: Chen, Yawen, et al.
Published: (2025)

The Multilingual Alignment Prism: Aligning Global and Local Preferences to Reduce Harm
by: Aakanksha, et al.
Published: (2024)