Episode
Predicting Decisions of AI Agents from Limited Interaction through Text-Tabular Modeling
- Podcast
- Daily Paper Cast
- Published
- May 15, 2026
- Duration seconds
- 1499
- Processing state
not_requested- Canonical source
- https://share.transistor.fm/s/3a7cb92e
Actions
POST https://stenobird.com/v1/public/podcasts/daily-paper-cast-7079649/episodes/predicting-decisions-of-ai-agents-from-limited-interaction-through-text-tabular-modeling/transcription-requests
Idempotently request low-priority transcript generation for this episode.GET https://stenobird.com/podcast/daily-paper-cast-7079649/predicting-decisions-of-ai-agents-from-limited-interaction-through-text-tabular-modeling.md
Read the agent-friendly Markdown representation of this episode resource.
Summary
🤗 Upvotes: 43 | cs.LG, cs.AI, cs.CL, cs.MA Authors: Eilam Shapira, Moshe Tennenholtz, Roi Reichart Title: Predicting Decisions of AI Agents from Limited Interaction through Text-Tabular Modeling Arxiv: http://arxiv.org/abs/2605.12411v1 Abstract: AI agents negotiate and transact in natural language with unfamiliar counterparts: a buyer bot facing an unknown seller, or a procurement assistant negotiating with a supplier. In such interactions, the counterpart's LLM, prompts, control logic, and rule-based fallbacks are hidden, while each decision can have monetary consequences. We ask whether an agent can predict an unfamiliar counterpart's next decision from a few interactions. To avoid real-world logging confounds, we study this problem in controlled bargaining and negotiation games, formulating it as target-adaptive text-tabular prediction: each decision point is a table row combining structured game state, offer history, and dialogue, while $K$ previous games of the same target agent, i.e., the counterpart being modeled, are provided in the prompt as labeled adaptation examples. Our model is built on a tabular foundation model that represents rows using game-state features and LLM-based text representations, and adds LLM-as-Observer as an additional representation: a small frozen LLM reads the decision-time state and dialogue; its answer is discarded, and its hidden state becomes a decision-oriented feature, making the LLM an encoder rather than a direct few-shot predictor. Training on 13 frontier-LLM agents and testing on 91 held-out scaffolded agents, the full model outperforms direct LLM-as-Predictor prompting and game+text features baselines. Within this tabular model, Observer features contribute beyond the other feature schemes: at $K=16$, they improve response-p…