Cover Image

Suppressing Pink Elephants with Direct Principle Feedback

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Castricato, Louis, Lile, Nathan, Anand, Suraj, Schoelkopf, Hailey, Verma, Siddharth, Biderman, Stella
Format:	Preprint
Published:	2024
Subjects:	Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Self-Directed Synthetic Dialogues and Revisions Technical Report
by: Lambert, Nathan, et al.
Published: (2024)

PERSONA: A Reproducible Testbed for Pluralistic Alignment
by: Castricato, Louis, et al.
Published: (2024)

PolyPythias: Stability and Outliers across Fifty Language Model Pre-Training Runs
by: van der Wal, Oskar, et al.
Published: (2025)

Negation: A Pink Elephant in the Large Language Models' Room?
by: Vrabcová, Tereza, et al.
Published: (2025)

Why Has Predicting Downstream Capabilities of Frontier AI Models with Scale Remained Elusive?
by: Schaeffer, Rylan, et al.
Published: (2024)

Llemma: An Open Language Model For Mathematics
by: Azerbayev, Zhangir, et al.
Published: (2023)

Big-Math: A Large-Scale, High-Quality Math Dataset for Reinforcement Learning in Language Models
by: Albalak, Alon, et al.
Published: (2025)

LLM Circuit Analyses Are Consistent Across Training and Scale
by: Tigges, Curt, et al.
Published: (2024)

Explaining and Mitigating Crosslingual Tokenizer Inequities
by: Arnett, Catherine, et al.
Published: (2025)

Towards System 2 Reasoning in LLMs: Learning How to Think With Meta Chain-of-Thought
by: Xiang, Violet, et al.
Published: (2025)

Transformer-Based Models Are Not Yet Perfect At Learning to Emulate Structural Recursion
by: Zhang, Dylan, et al.
Published: (2024)

Are PPO-ed Language Models Hackable?
by: Anand, Suraj, et al.
Published: (2024)

From Decoding to Meta-Generation: Inference-time Algorithms for Large Language Models
by: Welleck, Sean, et al.
Published: (2024)

LEACE: Perfect linear concept erasure in closed form
by: Belrose, Nora, et al.
Published: (2023)

Cognitive Behaviors that Enable Self-Improving Reasoners, or, Four Habits of Highly Effective STaRs
by: Gandhi, Kanishk, et al.
Published: (2025)

Social Choice Should Guide AI Alignment in Dealing with Diverse Human Feedback
by: Conitzer, Vincent, et al.
Published: (2024)

Dual Process Learning: Controlling Use of In-Context vs. In-Weights Strategies with Weight Forgetting
by: Anand, Suraj, et al.
Published: (2024)

KMMLU: Measuring Massive Multitask Language Understanding in Korean
by: Son, Guijin, et al.
Published: (2024)

The Responsible Foundation Model Development Cheatsheet: A Review of Tools & Resources
by: Longpre, Shayne, et al.
Published: (2024)

Leveraging LLM For Synchronizing Information Across Multilingual Tables
by: Khincha, Siddharth, et al.
Published: (2025)

The Elephant in the Room: Analyzing the Presence of Big Tech in Natural Language Processing Research
by: Abdalla, Mohamed, et al.
Published: (2023)

The Elephant in the Coreference Room: Resolving Coreference in Full-Length French Fiction Works
by: Bourgois, Antoine, et al.
Published: (2025)

Through the LLM Looking Glass: A Socratic Probing of Donkeys, Elephants, and Markets
by: Kennedy, Molly, et al.
Published: (2025)

When AI Co-Scientists Fail: SPOT-a Benchmark for Automated Verification of Scientific Research
by: Son, Guijin, et al.
Published: (2025)

Lessons from the Trenches on Reproducible Evaluation of Language Models
by: Biderman, Stella, et al.
Published: (2024)

Still Fresh? Evaluating Temporal Drift in Retrieval Benchmarks
by: Kuissi, Nathan, et al.
Published: (2026)

The Ghost in the Keys: A Disklavier Demo for Human-AI Musical Co-Creativity
by: Bradshaw, Louis, et al.
Published: (2025)

Elephant in the Room: Unveiling the Impact of Reward Model Quality in Alignment
by: Liu, Yan, et al.
Published: (2024)

InfoGatherer: Principled Information Seeking via Evidence Retrieval and Strategic Questioning
by: Taranukhin, Maksym, et al.
Published: (2026)

Exposing Pink Slime Journalism: Linguistic Signatures and Robust Detection Against LLM-Generated Threats
by: Shahriar, Sadat, et al.
Published: (2025)

$\forall$uto$\exists$val: Autonomous Assessment of LLMs in Formal Synthesis and Interpretation Tasks
by: Karia, Rushang, et al.
Published: (2024)

The Importance of Directional Feedback for LLM-based Optimizers
by: Nie, Allen, et al.
Published: (2024)

How English Print Media Frames Human-Elephant Conflicts in India
by: Punith, Bonala Sai, et al.
Published: (2026)

Elephants Never Forget: Testing Language Models for Memorization of Tabular Data
by: Bordt, Sebastian, et al.
Published: (2024)

Blind Men and the Elephant: Diverse Perspectives on Gender Stereotypes in Benchmark Datasets
by: Zakizadeh, Mahdi, et al.
Published: (2025)

Inducing Robustness in a 2 Dimensional Direct Preference Optimization Paradigm
by: Shashidhar, Sarvesh, et al.
Published: (2025)

Teaching LLMs to Abstain across Languages via Multilingual Feedback
by: Feng, Shangbin, et al.
Published: (2024)

Semantic Search Evaluation
by: Zheng, Chujie, et al.
Published: (2024)

Prismatic VLMs: Investigating the Design Space of Visually-Conditioned Language Models
by: Karamcheti, Siddharth, et al.
Published: (2024)

A Critical Study of What Code-LLMs (Do Not) Learn
by: Anand, Abhinav, et al.
Published: (2024)