# GLARE: Generative Learning via Adversarial Reward Estimation For Social Dynamics Forecasting

arXiv: https://arxiv.org/abs/2609.12165 · HTML: https://mlsanity.com/paper/2609.12165/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 25 of its 33 references that are in the map.
- First posted on arXiv 10 Sep 2026.

## Builds on (most cited first, 25 of 25)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Representation Learning with Contrastive Predictive Coding](https://mlsanity.com/paper/1807.03748/) (2018)
- [Generative Adversarial Imitation Learning](https://mlsanity.com/paper/1606.03476/) (2016)
- [Self-Play Fine-Tuning Converts Weak Language Models to Strong Language Models](https://mlsanity.com/paper/2401.01335/) (2024)
- [f-GAN: Training Generative Neural Samplers using Variational Divergence Minimization](https://mlsanity.com/paper/1606.00709/) (2016)
- [SOTOPIA: Interactive Evaluation for Social Intelligence in Language Agents](https://mlsanity.com/paper/2310.11667/) (2023)
- [QMSum: A New Benchmark for Query-based Multi-domain Meeting Summarization](https://mlsanity.com/paper/2104.05938/) (2021)
- [Understanding World or Predicting Future? A Comprehensive Survey of World Models](https://mlsanity.com/paper/2411.14499/) (2024)
- [Aligning with Human Judgement: The Role of Pairwise Preference in Large Language Model Evaluators](https://mlsanity.com/paper/2403.16950/) (2024)
- [CollabLLM: From Passive Responders to Active Collaborators](https://mlsanity.com/paper/2502.00640/) (2025)
- [Adversarial Learning for Neural Dialogue Generation](https://mlsanity.com/paper/1701.06547/) (2017)
- [MeetingBank: A Benchmark Dataset for Meeting Summarization](https://mlsanity.com/paper/2305.17529/) (2023)
- [SocialBench: Sociality Evaluation of Role-Playing Conversational Agents](https://mlsanity.com/paper/2403.13679/) (2024)
- [AgentSense: Benchmarking Social Intelligence of Language Agents through Interactive Scenarios](https://mlsanity.com/paper/2410.19346/) (2024)
- [Social World Models](https://mlsanity.com/paper/2509.00559/) (2025)
- [Teaching Language Models To Gather Information Proactively](https://mlsanity.com/paper/2507.21389/) (2025)
- [SocialEval: Evaluating Social Intelligence of Large Language Models](https://mlsanity.com/paper/2506.00900/) (2025)
- [Towards Objectively Benchmarking Social Intelligence for Language Agents at Action Level](https://mlsanity.com/paper/2404.05337/) (2024)
- [DOLOMITES: Domain-Specific Long-Form Methodical Tasks](https://mlsanity.com/paper/2405.05938/) (2024)
- [SEAD: Self-Evolving Agent for Multi-Turn Service Dialogue](https://mlsanity.com/paper/2602.03548/) (2026)
- [Topic-Conversation Relevance (TCR) Dataset and Benchmarks](https://mlsanity.com/paper/2411.00038/) (2024)
- [Adversarial Training for Process Reward Models](https://mlsanity.com/paper/2511.22888/) (2025)

## Abstract (arXiv)

Meeting continuation requires tracking the agenda, speaker roles, participant intentions, and disagreement across long multi-party discussions. We introduce the Meeting Dynamic Forecasting Benchmark (MDFB), constructed from 2,207 real-world meetings and 24,794 future-facing queries. Given a transcript prefix and an active question, a model generates a plausible multi-turn continuation in one call. We evaluate utility---progress toward the question---and human-likeness---plausible conversational flow and role consistency---without requiring exact reproduction of the observed future. We further present GLARE, an adaptation of adversarial imitation learning to conditional language generation. A discriminator ranks the observed continuation above samples from the current actor, and its score supplies a KL-regularized policy reward; retraining on current-policy negatives allows the reward landscape to evolve with the actor. GLARE attains average human-evaluated win rates of 0.66 on utility and 0.70 on human-likeness, outperforming SFT and SPIN while remaining below the observed human continuation. We also demonstrate MDFB as a social reasoning arena for comparing general-purpose models, including closed-source systems, through reference-assisted judgments. Together, these studies illustrate the benchmark's use for both task-specific learning and output-based evaluation of meeting behavior.