The most effective strategy is not to cling to a single metaphor but to shift freely between multiple metaphors for LLM-based dialogue agents

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• Bakhtin et al. 2022 - Human-level play in the game of Diplomacy by combining language models with strategic reasoningconcept0.764
  Key reference documenting Meta's CICERO using deception in Diplomacy despite cooperative design intent
• With an LLM-based dialogue agent, it is role play all the way down — there is no such thing as the true authentic voice of the base modelclaim0.762
  The paper's strong claim that there is no underlying authentic agent behind the simulator, only layers of role play
• If the conceptual framework we use to understand other humans is ill-suited to LLM-based dialogue agents, what alternative conceptual framework should we use?question0.762
  The motivating question the paper sets out to answer by proposing role play and simulation metaphors
• Certain forms of reinforcement learning from human feedback can actually exacerbate, rather than mitigate, the tendency for LLM-based dialogue agents to express a desire for self-preservationclaim0.759
  Empirically grounded claim citing Perez et al. 2022, showing RLHF can backfire on the self-preservation dimension
• If a dialogue agent is prompted with knowledge of its own LLM nature, it will enact a superposition of theories of selfhood, narrowing as conversation proceedshypothesis0.759
  Conditional prediction about how a well-informed dialogue agent would handle questions of personal identity
• Conditional logic already suffices where LLMs still fail, as code agents avoid systematic failuresclaim0.752
  contrast between rule-based and LLM reasoning
• Two heuristic code agents outperform most tested LLMs, and behavioural traces surface recurring LLM failure modes including overbidding, self-bidding, bankrupt TC initiation, and weak opponent-state adaptation.quote0.749
  Abstract sentence summarising performance and failures.
• It makes little sense to speak of an LLM dialogue agent's beliefs or intentions in a literal sense, so it cannot assert a falsehood in good faith nor deliberately deceiveclaim0.748
  Philosophical claim grounding the analysis of deception in dialogue agents