Cover Image

Alignment Tipping Process: How Self-Evolution Pushes LLM Agents Off the Rails

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Han, Siwei, Xiong, Kaiwen, Liu, Jiaqi, Ye, Xinyu, Su, Yaofeng, Duan, Wenbo, Liu, Xinyuan, Xie, Cihang, Bansal, Mohit, Ding, Mingyu, Zhang, Linjun, Yao, Huaxiu
Format:	Preprint
Published:	2025
Subjects:	Machine Learning Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

SimpleMem: Efficient Lifelong Memory for LLM Agents
by: Liu, Jiaqi, et al.
Published: (2026)

ReAgent-V: A Reward-Driven Multi-Agent Framework for Video Understanding
by: Zhou, Yiyang, et al.
Published: (2025)

EvolveMem:Self-Evolving Memory Architecture via AutoResearch for LLM Agents
by: Liu, Jiaqi, et al.
Published: (2026)

Agent0-VL: Exploring Self-Evolving Agent for Tool-Integrated Vision-Language Reasoning
by: Liu, Jiaqi, et al.
Published: (2025)

MetaClaw: Just Talk -- An Agent That Meta-Learns and Evolves in the Wild
by: Xia, Peng, et al.
Published: (2026)

When and How Much to Imagine: Adaptive Test-Time Scaling with World Models for Visual Spatial Reasoning
by: Yu, Shoubin, et al.
Published: (2026)

ClawArena: Benchmarking AI Agents in Evolving Information Environments
by: Ji, Haonian, et al.
Published: (2026)

Multimodal Representation Learning by Alternating Unimodal Adaptation
by: Zhang, Xiaohui, et al.
Published: (2023)

Omni-SimpleMem: Autoresearch-Guided Discovery of Lifelong Multimodal Agent Memory
by: Liu, Jiaqi, et al.
Published: (2026)

Analyzing and Mitigating Object Hallucination in Large Vision-Language Models
by: Zhou, Yiyang, et al.
Published: (2023)

On Track or Off The Rails?
by: Rademann, Jenny
Published: (2022)

ClawForge: Generating Executable Interactive Benchmarks for Command-Line Agents
by: Lai, Yuxiang, et al.
Published: (2026)

SAFREE: Training-Free and Adaptive Guard for Safe Text-to-Image And Video Generation
by: Yoon, Jaehong, et al.
Published: (2024)

Polynomial escape rates via maximal large deviations
by: Su, Yaofeng
Published: (2024)

Knowing the Answer Isn't Enough: Fixing Reasoning Path Failures in LVLMs
by: Wang, Chaoyang, et al.
Published: (2025)

AQA-Bench: An Interactive Benchmark for Evaluating LLMs' Sequential Reasoning Ability
by: Yang, Siwei, et al.
Published: (2024)

GRAPE: Generalizing Robot Policy via Preference Alignment
by: Zhang, Zijian, et al.
Published: (2024)

When Infrastructure Pushes Residents Out: High‐Speed Rail and Tourism Outflows
by: David Boto‐García, et al.
Published: (2026)

SkillRL: Evolving Agents via Recursive Skill-Augmented Reinforcement Learning
by: Xia, Peng, et al.
Published: (2026)

Distribution-Free Fair Federated Learning with Small Samples
by: Yin, Qichuan, et al.
Published: (2024)

Back to Boundaries in Billiards
by: Bunimovich, Leonid, et al.
Published: (2022)

S$^{2}$FT: Efficient, Scalable and Generalizable LLM Fine-tuning by Structured Sparsity
by: Yang, Xinyu, et al.
Published: (2024)

MPO: An Efficient Post-Processing Framework for Mixing Diverse Preference Alignment
by: Wang, Tianze, et al.
Published: (2025)

Which subsets and when orbits of non-uniformly hyperbolic systems prefer to visit: operator renewal theory approach
by: Bunimovich, Leonid A., et al.
Published: (2024)

Statistics of Weakly Chaotic Systems
by: Bunimovich, Leonid A., et al.
Published: (2025)

Efficient and Sharp Off-Policy Evaluation in Robust Markov Decision Processes
by: Bennett, Andrew, et al.
Published: (2024)

SimpleOCR: Rendering Visualized Questions to Teach MLLMs to Read
by: Peng, Yibo, et al.
Published: (2026)

Conformal Prediction for Deep Classifier via Label Ranking
by: Huang, Jianguo, et al.
Published: (2023)

FactTest: Factuality Testing in Large Language Models with Finite-Sample and Distribution-Free Guarantees
by: Nie, Fan, et al.
Published: (2024)

Epitaxial Lift‐Off Process for Triple‐Junction Solar Cells Using Internal Push–Pull Stressors
by: Prabudeva Ramu, et al.
Published: (2025)

Rethinking Interactive Image Segmentation with Low Latency, High Quality, and Diverse Prompts
by: Liu, Qin, et al.
Published: (2024)

PEANuT: Parameter-Efficient Adaptation with Weight-aware Neural Tweakers
by: Zhong, Yibo, et al.
Published: (2024)

Multi-state Modeling of Delay Evolution in Suburban Rail Transports
by: Colombo, Stefania, et al.
Published: (2025)

GPT-IMAGE-EDIT-1.5M: A Million-Scale, GPT-Generated Image Dataset
by: Wang, Yuhan, et al.
Published: (2025)

See It from My Perspective: How Language Affects Cultural Bias in Image Understanding
by: Ananthram, Amith, et al.
Published: (2024)

Self‐Supporting CoMo/ME‐ITO/SCG Catalyst Derived From Spent Coffee Grounds for Highly Efficient Hydrogen Evolution in Both Acidic and Alkaline Media
by: Kaiwen Liu, et al.
Published: (2026)

The Power of Summary-Source Alignments
by: Ernst, Ori, et al.
Published: (2024)

Disperse-Then-Merge: Pushing the Limits of Instruction Tuning via Alignment Tax Reduction
by: Fu, Tingchen, et al.
Published: (2024)

Transform Before You Query: A Privacy-Preserving Approach for Vector Retrieval with Embedding Space Alignment
by: He, Ruiqi, et al.
Published: (2025)

Agent0: Unleashing Self-Evolving Agents from Zero Data via Tool-Integrated Reasoning
by: Xia, Peng, et al.
Published: (2025)