Salesforce AI Research

Looking for the cutting-edge of AI research? Follow Salesforce AI Research to see how we're transforming enterprise technology through advanced innovations. From world models to agentic systems, discover the future of AI before it hits the market.

10/ That is VibeTrain: AI model and agent training as a conversation.

9/ The thesis: As AI moves from model training to agent training, the workflow needs an agent too. Not just a dashboard. Not just an API. A collaborator that can reason, code, train, diagnose, compare, report, and deploy.

1/ What if you could kick off an RL training run with just a broken AI agent — no dataset to prepare, no rollout code to write? Meet SFR-VibeTrain, built at @Salesforce AI Research: a @SlackHQ-native training agent on top of our large-scale training stack, SFR-RL. In the demo, a customer-retention agent is failing. VibeTrain turns it into synthetic scenarios, a simulator, a judge, training, evaluation, deployment, and a live sandbox. #AgenticAI #RL #LLM #VibeTrain

A great moment for the lab! 🎉 "Don't Stop Early" earned its spot on the Industry Track for advancing enterprise deep research at scale. Controlled information flow and evidence-aware termination prevent premature stopping and uneven coverage. bit.ly/49fk2zQ #ACL2026

(4/4) 📄 MAS-ProVe: Understanding the Process Verification of Multi-Agent Systems Examining how multi-agent systems verify their own reasoning processes — a step toward more reliable agentic workflows. bit.ly/4bukYRp

(3/4) 📄 MAS-Orchestra: Understanding and Improving Multi-Agent Reasoning Through Holistic Orchestration and Controlled Benchmarks A framework + benchmark for studying how multi-agent systems reason — and how orchestration choices shape outcomes. bit.ly/4adooHA

🎉 3 papers from @Salesforce AI Research accepted to #ICML2026, spanning Mixture-of-Experts efficiency and multi-agent reasoning + verification. Learn more ↓

Correction: Congratulations to @PhilippeLaban!

Our Infrastructure & DevOps team was featured by @GoogleCloud. Lavanya Karanam and Avinash Gudagi on how @Salesforce is solving one of the hardest problems in large-scale AI training: keeping powerful GPUs fully fed. bit.ly/4n6HDc3 #FutureOfAI #EnterpriseAI

3 AI trends shaping #EnterpriseAI: simulation environments, ambient intelligence, and agent-to-agent communication. @Digitai79 on our focus on Enterprise General Intelligence (#EGI) in @gregorojstersek's #TDX26 recap. bit.ly/4mT4irW #FutureOfAI

Don't Stop Early: Scalable Enterprise Deep Research with Controlled Information Flow and Evidence-Aware Termination Paper: bit.ly/49fk2zQ Enterprise deep research often fails to produce decision-ready reports due to uneven coverage, context explosion, and premature stopping. This work introduces an EDR architecture built on three mechanisms: → 🗂️ Coverage-driven task decomposition via outline generation with reflection → 🔗 Dependency-guided execution with localized contexts, so information flows only when explicitly required → ✅ Evidence-based termination criteria that keep agents gathering until sufficiency conditions are met 🏆 Results: 82.09 HAA and 4.31 coverage on internal sales enablement, outperforming Gemini DR, OpenAI DR, ODR, and DeerFlow. Highly competitive in comprehensiveness (54.85) and insightfulness (55.63) on DeepResearch Bench. 👥 Authors: Prafulla Kumar Choubey, Kung-Hsiang Huang @steeve__huang, Pranav Narayanan Venkit, Jiaxin Zhang @jxzhangjhu, Vaibhav Vats, Yu Li, Xiangyu Peng, Chien-Sheng Wu @jasonwu0731 #FutureOfAI #EnterpriseAI #AgenticAI

Heading to the DATA-FM workshop at #ICLR2026? 🦾 Catch my invited talk at 1:30 PM today in Room 203A+B: "Agentic Ambient Intelligence: Perception, Reasoning & Action." I’ll be diving into 4 new research pillars for moving AI into the physical world. Come say hi! 📍