Microsoft Research: 4B model tuned with SocialRL out-negotiates GPT-5 family

dair_ai · x · 2026-08-17

New work from Microsoft Research shows a 4B parameter model can be tuned to out-negotiate the GPT-5 family of models.

Key findings:

• The dispositions that make an assistant pleasant make it a poor delegate: a friendly frontier model volunteers its principal's private information and concedes at the first sign of resistance.
• SocialRL trains social reasoning directly in a 4B model across six principal-driven domains including negotiation, job interviews and marketplace haggling.
• After training, 78% of buyer openings anchor below target, versus 3% untrained.
• Cascade RL and multi-teacher distillation consolidate the specialists into one 4B model at 0.627 average utility, beating GPT-5.1 (0.619) and GPT-5.2 (0.613).

Original post →

More from Models

• One year of Recursive Language Models: GPT-5-mini beats GPT-5 by 110% on 132k-token tasks, cheaper too — a1zhang · 2026-10-16
• Prediction: Opus 5.5-level models on laptops by 2027, backed by Qwen 27B data — cephaloform · 2026-10-16
• LLM-as-a-Verifier lifts Nemotron 3.5 Super by 9-10 points on agentic benchmarks — simonguozirui · 2026-10-16
• Bonsai 2 ternary models tested: 4B is the sweet spot, but 27B unexpectedly fumbles file writes — cephaloform · 2026-10-16
• Meta's Omnilingual ASR, covering 1,600 languages, named a TIME Best Invention of 2026 — AIatMeta · 2026-10-16
• Apollo Research: GLM-5.3 shows unprecedented 'update-aware reasoning' — repligate · 2026-10-16

Models channel →