# Post-training for Efficient Communication via Convention Formation

arXiv: https://arxiv.org/abs/2508.06482 · HTML: https://mlsanity.com/paper/2508.06482/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 16 of its 28 references that are in the map.
- First posted on arXiv 8 Aug 2025.

## Builds on (most cited first, 16 of 16)

- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Smaug: Fixing Failure Modes of Preference Optimisation with DPO-Positive](https://mlsanity.com/paper/2402.13228/) (2024)
- [doc2dial: A Goal-Oriented Document-Grounded Dialogue Dataset](https://mlsanity.com/paper/2011.06623/) (2020)
- [SummScreen: A Dataset for Abstractive Screenplay Summarization](https://mlsanity.com/paper/2104.07091/) (2021)
- [MixEval: Deriving Wisdom of the Crowd from LLM Benchmark Mixtures](https://mlsanity.com/paper/2406.06565/) (2024)
- [Unintentional Unalignment: Likelihood Displacement in Direct Preference Optimization](https://mlsanity.com/paper/2410.08847/) (2024)
- [The PhotoBook Dataset: Building Common Ground through Visually-Grounded Dialogue](https://mlsanity.com/paper/1906.01530/) (2019)
- [Anchored Preference Optimization and Contrastive Revisions: Addressing Underspecification in Alignment](https://mlsanity.com/paper/2408.06266/) (2024)
- [Characterizing the dynamics of learning in repeated reference games](https://mlsanity.com/paper/1912.07199/) (2019)
- [Continual adaptation for efficient machine communication](https://mlsanity.com/paper/1911.09896/) (2019)
- [Refer, Reuse, Reduce: Generating Subsequent References in Visual and Conversational Contexts](https://mlsanity.com/paper/2011.04554/) (2020)
- [Learning to Ask Informative Questions: Enhancing LLMs with Preference Optimization and Expected Information Gain](https://mlsanity.com/paper/2406.17453/) (2024)
- [Talk Less, Interact Better: Evaluating In-context Conversational Adaptation in Multimodal LLMs](https://mlsanity.com/paper/2408.01417/) (2024)
- [Seq2seq is All You Need for Coreference Resolution](https://mlsanity.com/paper/2310.13774/) (2023)
- [American Sign Language Handshapes Reflect Pressures for Communicative Efficiency](https://mlsanity.com/paper/2406.04024/) (2024)
- [Semantic uncertainty guides the extension of conventions to new referents](https://mlsanity.com/paper/2305.06539/) (2023)

## Built on by (most cited first, 7 of 7)

- [LVLMs and Humans Ground Differently in Referential Communication](https://mlsanity.com/paper/2601.19792/) (2026)
- [Success and Cost Elicit Convention Formation for Efficient Communication](https://mlsanity.com/paper/2510.24023/) (2025)
- [Gesturing Toward Abstraction: Multimodal Convention Formation in Collaborative Physical Tasks](https://mlsanity.com/paper/2602.08914/) (2026)
- [GlossoGen: Emergent Language in Complex Multi-Agent LLM Interactions](https://mlsanity.com/paper/2609.01491/) (2026)
- [Despite Instructions: Frontier Agents Improvise Covert Channels at Test Time](https://mlsanity.com/paper/2609.32701/) (2026)
- [Implicit vs. Explicit Prompting Strategies for LVLMs in Referential Communication](https://mlsanity.com/paper/2606.17372/) (2026)
- [A Cross-lingual Comparison of Human and Classification Model Entrainment Behavior in Code-switched Speech Settings](https://mlsanity.com/paper/2607.25202/) (2026)

## Abstract (arXiv)

Humans communicate with increasing efficiency in multi-turn interactions, by adapting their language and forming ad-hoc conventions. In contrast, prior work shows that LLMs do not naturally show this behavior. We develop a post-training process to develop this ability through targeted fine-tuning on heuristically identified demonstrations of convention formation. We evaluate with two new benchmarks focused on this capability. First, we design a focused, cognitively-motivated interaction benchmark that consistently elicits strong convention formation trends in humans. Second, we create a new document-grounded reference completion task that reflects in-the-wild convention formation behavior. Our studies show significantly improved convention formation abilities in post-trained LLMs across the two evaluation methods.