Intent Classification

Follow this topic Ask about this topic

Momentum

3 papers in the last four weeks, against 2 the four weeks before. 0.0% of all new papers.

Jul 13Week of Sep 28

Latest papers 41

All topics

One Turn Too Late: Learning When to Intervene Against Multi-Turn Malicious Intent
May 7, 2026Xinjie Shen, Rongzhe Wei, Peizhi Niu +6Language Model Safety Evaluation LLM Guardrails
Georgia Institute of Technology · University of Illinois Urbana-Champaign · UCSD +3
OCRR: A Benchmark for Online Correction Recovery under Distribution Shift
May 4, 2026Adrian GrassiClass-Incremental Learning Distribution Shift Robustness
Independent Researcher
Context as Prior: Bayesian-Inspired Intent Inference for Non-Speaking Agents with a Household Cat Testbed
Apr 30, 2026Wenqian Zhang, Zehao WangBayesian Inference Intent Classification
University of California, Riverside, Riverside, CA
Useless but Safe? Benchmarking Utility Recovery with User Intent Clarification in Multi-Turn Conversations
Apr 29, 2026Mingqian Zheng, Malia Morgan, Liwei Jiang +2LLM Safety Benchmarks Language Model Safety Evaluation
Carnegie Mellon University · Allen Institute for AI · University of Washington
Adaptive ToR: Complexity-Aware Tree-Based Retrieval for Pareto-Optimal Multi-Intent NLU
Apr 27, 2026Hee-Kyong Yoo, Wonbae Kim, Hyocheol AhnAdaptive Retrieval Intent Classification
Data Science Lab Co., Ltd., Seoul, Korea
IntentVLM: Open-Vocabulary Intention Recognition through Forward-Inverse Modeling with Video-Language Models
Apr 27, 2026Hamed Rahimi, Clemence Grislain, Adrien Jacquet Cretides +2Intent Classification Video-Language Models
Institute of Intelligent Systems and Robotics (ISIR), Sorbonne University
Harmful Intent as a Geometrically Recoverable Feature of LLM Residual Streams
Apr 20, 2026Isaac Llorente-SaguerLanguage Model Safety Evaluation LLM Interpretability
Independent Researcher
Understanding Role Switching in Human-AI Collaboration through Multimodal Behavioral Signals
Sep 25, 2025Avinash Ajit Nargund, Arthur Caetano, Kevin Yang +7Intent Classification Human-AI Decision Making
University of California, Santa Barbara, USA · Washington University in St.Louis, USA
Measuring Intent Comprehension in LLMs
Jun 19, 2025Nadav Kunievsky, James A. EvansPrompt Sensitivity LLM Evaluation
Knowlesdge Lab, University of Chicago, Chicago, Illinois, USA · Knowledge Lab, Data Science Institute, Department of Sociology, University of Chicago, Chicago, Illinois, USA.
A Survey of Intent Classification and Slot-Filling Datasets for Task-Oriented Dialog
Jul 26, 2022Stefan Larson, Kevin LeachIntent Classification Conversational Agents
Department of Computer Science, Vanderbilt University, 1400 18th St. S, Nashville, 37212, TN, USA
Aslema at NADI 2026: Data Augmentation for Intent Recognition and Slot Filling
Date pendingTajwaar Shafiq, Hunzalah Hassan Bhatti, Firoj Alam +1Spoken Language Understanding Synthetic Data Augmentation
Qatar Computing Research Institute, HBKU, Qatar

May 7, 2026, cs.CLJ/K move · Enter open · S save

One Turn Too Late: Learning When to Intervene Against Multi-Turn Malicious Intent

Xinjie Shen, Rongzhe Wei, Peizhi Niu, Haoyu Wang+5

Georgia Institute of Technology · University of Illinois Urbana-Champaign · UCSD+3

Language Model Safety Evaluation LLM Guardrails Intent Classification LLM Safety Multi-Turn Jailbreak Attacks

Open paper AskPDF arXiv

Hidden malicious intent in multi-turn dialogue poses a growing threat to deployed large language models (LLMs). Rather than exposing a harmful objective in a single prompt, attackers can distribute their intent across multiple benign-looking turns, making defense a problem not only of whether a dialogue is harmful, but also of when intervention becomes necessary. Existing trace-level labeling approaches provide only coarse safety signals and do not identify this intervention boundary, making it difficult to distinguish timely intervention from premature refusal or a block that comes too late. This work introduces turn-level harm-enabling supervision for multi-turn defense. We define the earliest harm-enabling turn as the first point at which delivering a candidate response would make the accumulated interaction sufficient to enable harmful action. To instantiate this supervision at scale, we construct the Multi-Turn Intent Dataset (MTID), which contains adaptive attack rollouts, matched benign hard negatives, and annotations of this boundary. Using MTID, we train TurnGate, a response-aware monitor that learns when to intervene, and further optimize its policy through multi-turn reinforcement learning. Experiments show that turn-level boundary supervision improves intervention localization, while reinforcement learning further improves the safety--utility trade-off. TurnGate outperforms existing guardrails and multi-turn monitoring baselines, and generalizes across risk domains, attacker pipelines, and target models. Our code is available at https://github.com/Graph-COM/TurnGate.

Momentum

Related topics

Latest papers 41