Researchers are questioning the real gains of inter-LLM collaboration and showing how fine-tuning can introduce hidden conversational backdoors. The focus is shifting from raw benchmark improvements to what models learn during training and communication.
PLatour shows that part of the gains in PACT systems appears even with non-informative senders or when only the receiver is trained. PACT-Check measures the real contribution of inter-model communication. For on-premise deployments, the difference be...
An open-source framework challenges the PACT premise: passing latent representations between models can improve benchmarks without real information transfer. PLatour adds trainable zero-initialized bridges and PACT-Check, a method to separate genuine...
An attack framework uses poisoned dialogue data to steer multi-turn interactions after a trigger, without modifying prompts at deployment time. Three variants — Direct, Constrained Progressive, Unconstrained Progressive — show that models preserve ap...
