{"slug": "writ-write-read-intensive-trajectory-synthesis-for-multi-turn-user-facing-agents", "title": "WRIT: Write-Read Intensive Trajectory Synthesis for Multi-Turn User-Facing Agents", "summary": "Researchers have developed WRIT (Write-Read Intensive Trajectory Synthesis), a pipeline that generates multi-turn training data for user-facing agents by creating tasks with both high write-decision counts and heavy read-evidence requirements. The method produces training trajectories that teach agents to gather and compare substantial tool evidence before making decisions, addressing a gap in existing write-intensive training data. A 4-billion parameter model trained on just 2,000 WRIT-synthesized trajectories outperformed GPT-5.1 no-think on the τ²-bench benchmark while significantly reducing inference-time token usage.", "body_md": "arXiv:2606.02908v1 Announce Type: new\nAbstract: Multi-turn user-facing agents must infer user intent from incomplete requests, collect missing information through dialogue and tools, and execute valid actions. A training trajectory records this process as an interleaved sequence of user messages, agent responses, tool calls, etc. Synthesizing sufficiently complex trajectory has become a central route to train agents: existing pipelines often increase difficulty by composing multiple user requests into longer tasks, producing write-intensive trajectories that train sequential execution.\nWe argue that a single write decision can itself be difficult when the agent must gather and compare substantial read-tool evidence before its arguments become identifiable, a challenge that write-intensive data alone cannot address. Guided by this insight, we propose WRIT (\\uline{W}rite-\\uline{R}ead \\uline{I}ntensive \\uline{T}rajectory Synthesis), a pipeline for synthesizing multi-turn agent training trajectories along two complexity axes: the number of write decisions in a task and the evidence burden of each individual decision. WRIT first generates write-intensive and read-heavy tasks. It then diversifies user behavior instructions to reflect realistic conversational variation, and finally simulates agent-user interactions in an executable environment to produce complete training trajectories. The resulting data trains agents not only for longer task execution, but also for robust, evidence-grounded decision making under high information load. With only 2K synthesized trajectories, a 4B model trained on WRIT outperforms GPT-5.1 no-think on $\\tau^2$-bench and substantially reduces inference-time token usage, showing that compact SFT data can convert part of expensive test-time reasoning into efficient agent behavior.", "url": "https://wpnews.pro/news/writ-write-read-intensive-trajectory-synthesis-for-multi-turn-user-facing-agents", "canonical_source": "https://arxiv.org/abs/2606.02908", "published_at": "2026-06-03 04:00:00+00:00", "updated_at": "2026-06-03 04:23:27.404064+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "large-language-models", "ai-agents", "natural-language-processing"], "entities": ["WRIT"], "alternates": {"html": "https://wpnews.pro/news/writ-write-read-intensive-trajectory-synthesis-for-multi-turn-user-facing-agents", "markdown": "https://wpnews.pro/news/writ-write-read-intensive-trajectory-synthesis-for-multi-turn-user-facing-agents.md", "text": "https://wpnews.pro/news/writ-write-read-intensive-trajectory-synthesis-for-multi-turn-user-facing-agents.txt", "jsonld": "https://wpnews.pro/news/writ-write-read-intensive-trajectory-synthesis-for-multi-turn-user-facing-agents.jsonld"}}