Beyond Small Talk: Modeling Disclosure Trajectories in Human-AI Conversation
We thus introduce Trajectory-based Disclosure Evaluation (TraDE), an evaluation framework grounded in Social Penetration Theory that utilizes predefined disclosure trajectories and evaluates how preceding disclosure history shapes model behavior under matched current disclosure.
ProofPaper ↗
Key points
- Recently, we see a rise in conversational agents as a form of social and emotional support, where disclosure will unfold over time.
- Yet, current evaluations of these agents typically examine isolated conversational states.
- Through the use of TraDE, we demonstrate across multiple models and relational behavior benchmarks that responses to disclosures matched on depth can differ depending on the preceding trajectory, particularly in whether models accept or challenge the user's framing.
- For several behaviors, we also demonstrate that models diverge most consistently from human supporters in the Emotional Support Conversation corpus (ESConv) not when users open up but rather when they step back in their disclosure.
Sources (1)
- [1]Beyond Small Talk: Modeling Disclosure Trajectories in Human-AI ConversationarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 21, 08:36 AM
We thus introduce Trajectory-based Disclosure Evaluation (TraDE), an evaluation framework grounded in Social Penetration Theory that utilizes predefined disclosure trajectories and evaluates how preceding disclosure history shapes model behavior under matched current disclosure.
Recently, we see a rise in conversational agents as a form of social and emotional support, where disclosure will unfold over time.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 21, 2026From VLM Planners to Robotic Agents: Self-Evolving Harnesses for History-Dependent Manipulation
- Oct 20, 2026Beyond the KV cache: Training and serving i14’s Laplace 3B linear attention model on AWS Trainium2 with native PyTorch
- Oct 20, 2026Voxtral 3
- Oct 20, 2026Do Diffusion LLMs Scale? Testing Mercury 3 + Decide
- Oct 20, 2026OverSeer: Multi-Agent World Models Beyond What Agents See
- Oct 19, 2026How Deepgram optimized voice AI at scale with AWS Trainium