mlsanityData to 7 Oct 2026

Towards Safety and Helpfulness Balanced Responses via Controllable Large Language Models

arXiv 2404.01295 ↗First posted 1 Apr 2024cs.CL cs.AIcited by 14

Not on the map. Why?

Abstract (arXiv)

As large language models (LLMs) become easily accessible nowadays, the trade-off between safety and helpfulness can significantly impact user experience. A model that prioritizes safety will cause users to feel less engaged and assisted while prioritizing helpfulness will potentially cause harm. Possible harms include teaching people how to build a bomb, exposing youth to inappropriate content, and hurting users' mental health. In this work, we propose to balance safety and helpfulness in diverse use cases by controlling both attributes in LLM. We explore training-free and fine-tuning methods that do not require extra human annotations and analyze the challenges of controlling safety and helpfulness in LLMs. Our experiments demonstrate that our method can rewind a learned model and unlock its controllability.

Key facts. Cited by 14 papers in the map (data through 7 Oct 2026). Builds on 25 of its 34 references that are in the map. First posted on arXiv 1 Apr 2024.

Builds on

25 of 34 references.

References · 25

  1. 2022Training language models to follow instructions with human feedbackcited by 15,712
  2. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,491
  3. 2017Proximal Policy Optimization Algorithmscited by 14,441
  4. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 13,958
  5. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 7,916
  6. 2019BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehensioncited by 5,874
  7. 2022PaLM: Scaling Language Modeling with Pathwayscited by 5,304
  8. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,040
  9. 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,634
  10. 2014Sequence to Sequence Learning with Neural Networkscited by 2,974
  11. 2019The Curious Case of Neural Text Degenerationcited by 2,881
  12. 2019Fine-Tuning Language Models from Human Preferencescited by 2,124
13 more
  1. 2019Language Models as Knowledge Bases?cited by 1,922
  2. 2018Personalizing Dialogue Agents: I have a dog, do you have pets too?cited by 820
  3. 2019Plug and Play Language Models: A Simple Approach to Controlled Text Generationcited by 793
  4. 2023RLAIF vs. RLHF: Scaling Reinforcement Learning from Human Feedback with AI Feedbackcited by 615
  5. 2015Sequence Level Training with Recurrent Neural Networkscited by 568
  6. 2016A Persona-Based Neural Conversation Modelcited by 221
  7. 2023Fine-tuning Language Models for Factualitycited by 219
  8. 2022RankGen: Improving Text Generation with Large Ranking Modelscited by 72
  9. 2017A causal framework for explaining the predictions of black-box sequence-to-sequence modelscited by 58
  10. 2020Controlling Style in Generated Dialoguecited by 38
  11. 2018Proximal Policy Optimization and its Dynamic Version for Sequence Generationcited by 7
  12. 2021Local Explanation of Dialogue Response Generationcited by 7
  13. 2018Improving Conditional Sequence Generative Adversarial Networks by Stepwise Evaluationcited by 6

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 4; 2025: 7; 2026: 3.

14 papers.

Sort:

By area:

  1. 2024OR-Bench: An Over-Refusal Benchmark for Large Language Modelscited by 222
  2. 2025Advances and Challenges in Foundation Agents: From Brain-Inspired Intelligence to Evolutionary, Collaborative, and Safe Systemscited by 125
  3. 2025Reasoning as an Adaptive Defense for Safetycited by 22
  4. 2024Jailbreak Antidote: Runtime Safety-Utility Balance via Sparse Representation Adjustment in Large Language Modelscited by 21
  5. 2024You Know What I'm Saying: Jailbreak Attack via Implicit Referencecited by 13
  6. 2025EVOREFUSE: Evolutionary Prompt Optimization for Evaluation and Mitigation of LLM Over-Refusal to Pseudo-Malicious Instructionscited by 6
  7. 2025Position: Beyond Assistance -- Reimagining LLMs as Ethical and Adaptive Co-Creators in Mental Health Carecited by 3
  8. 2026ForesightSafety Bench: A Frontier Risk Evaluation and Governance Framework towards Safe AIcited by 3
  9. 2025Accelerating RLHF Training with Reward Variance Increasecited by 2
  10. 2025High Accuracy, Less Talk (HALT): Reliable LLMs through Capability-Aligned Finetuningcited by 2
  11. 2025How Semantically Stable Are LLM Refusals? Measuring Confusion in Local Safety Boundariescited by 2
  12. 2024A Gradient Analysis Framework for Rewarding Good and Penalizing Bad Examples in Language Modelscited by 1
2 more
  1. 2026Hidden in the Request: Explaining Unethical LLM Compliance through Token Relevancecited by 1
  2. 2026Evolving and Detecting Multi-Turn Deception using Geometric Signaturescited by 0