Search Results for author:
Found 2 papers, 2 papers with code
Emotion-Anchored Contrastive Learning Framework for Emotion Recognition in Conversation
To achieve this, we utilize label encodings as anchors to guide the learning of utterance representations and design an auxiliary loss to ensure the effective separation of anchors for similar emotions.
Ranked #5 on
Emotion Recognition in Conversation
on MELD
COLD-Attack: Jailbreaking LLMs with Stealthiness and Controllability
Based on this connection, we adapt the Energy-based Constrained Decoding with Langevin Dynamics (COLD), a state-of-the-art, highly efficient algorithm in controllable text generation, and introduce the COLD-Attack framework which unifies and automates the search of adversarial LLM attacks under a variety of control requirements such as fluency, stealthiness, sentiment, and left-right-coherence.
