analiza aiqbench
This arXiv preprint studies how two LLMs converge on agreement across multi-turn debates, using a linear probe on internal activations. Its main finding is that agreement is hard to read from single conversations but shows Markov-like structure and a power-law link to external behavior once aggregated across many conversations. For readers, the work is a useful interpretability result on multi-agent LLM dynamics, though it is an early arXiv preprint whose conclusions depend on the specific models and subjective topics tested.
Seeing the Forest for the Trees: Emergent Agreement from Individual Conversations
arXiv:2610.29328v1 Announce Type: new Abstract: As LLM agents interact in the wild and shape each other's decisions, understanding how they converge toward agreement becomes important. We study this in multi-turn conversations where two LLMs discuss subjective matters, can agree or disagree, and may change their stances. Using a linear classifier on individual conversation activations, we identify an internal representation of agreement. We find that it varies substantially across turns and conversations and does not reliably track external agreement at the individual conversation level. We therefore move beyond the representation of agreement in individual conversations, by aggregating the agreement representation across conversations. Externally, this aggregate view reveals a simple structure: when pooled across conversations, agreement can be predicted well with a Markov process, with agreement at each turn being a direct function of agreement at the previous turn. Similarly, when internal agreement is aggregated, it couples with the external Markov agreement through a power-law mapping, exposing Markovian dynamics in internal representations. Additionally, we find that after a change of stance, models express agreement externally before their internal representations reflect the same shift. These results reveal a joint structure between internal and external agreement that emerges only when internal features are aggregated across conversations. Code available at https://github.com/technion-cs-nlp/SeeingTheForestForTheTrees.
Abstract:As LLM agents interact in the wild and shape each other's decisions, understanding how they converge toward agreement becomes important. We study this in multi-turn conversations where two LLMs discuss subjective matters, can agree or disagree, and may change their stances. Using a linear classifier on individual conversation activations, we identify an internal representation of agreement. We find that it varies substantially across turns and conversations and does not reliably track external agreement at the individual conversation level. We therefore move beyond the representation of agreement in individual conversations, by aggregating the agreement representation across conversations. Externally, this aggregate view reveals a simple structure: when pooled across conversations, agreement can be predicted well with a Markov process, with agreement at each turn being a direct function of agreement at the previous turn. Similarly, when internal agreement is aggregated, it couples with the external Markov agreement through a power-law mapping, exposing Markovian dynamics in internal representations. Additionally, we find that after a change of stance, models express agreement externally before their internal representations reflect the same shift. These results reveal a joint structure between internal and external agreement that emerges only when internal features are aggregated across conversations. Code available at this https URL.
Computation and Language (cs.CL)
arXiv:2610.29328 [cs.CL]
(or arXiv:2610.29328v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2610.29328 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 298 cuvinte
Subiecte conexe
How Conversational AI Chatbots Work in 2026
Slack Blog · 28 sept. 2026
Automatic Sentence Completion Using Bengali Audio Speech with a Transformer-Based Early Speech Recognition System
arXiv · 16 oct. 2026
The user in the LLM might be a moral patient too (a strange implication of the PSM and simulators)
LessWrong · 15 oct. 2026
Which character are we evaluating? Persona stability and AI welfare
LessWrong · 11 sept. 2026