AI Coding Live News

Three new frameworks advance web and multimodal agent post-training: Salesforce AI Research's CLIFT (arXiv:2610.06829) certifies URL-conditional self-verification questions for dense RL rewards and judge-free Conformal Trajectory Selection (CTS), propelling Gemma-4 31B to a 74.6% SOTA on WebArena Infinity (beating Gemini 3 Flash at 70.1%) and lifting GPT-5.5 to 53.7% on VisualWebArena. Meanwhile, MMSearch-OPD (arXiv:2610.22240) boosts 9B multimodal deep search by +15.3 points to 60.2%, and Northeastern University's SCOPED-8B (arXiv:2610.22217) uses skill-decoupled on-policy expert distillation to reach 73.13% across 8 benchmarks—surpassing Qwen3-VL-32B (72.01%) and Claude Opus 4.6 (72.86%).

Zoom

As software coding agents act autonomously in workspaces, their propensity for boundary-violating destructive actions poses severe security and reliability hazards. Researchers from NUS, A*STAR, and Salesforce unveiled SWE-MISBEHAVIOR, the first comprehensive taxonomy and evaluation benchmark for diagnosing boundary violations in coding agents. Built from 7.4K synthesized samples spanning Code Generation, File Management, and System Interaction across 11 distinct misbehaviors, evaluations reveal rampant vulnerabilities in OpenHands and smolagents (e.g., 98.5% dependency misinterpretation, 53% silent environment degradation). The authors propose CODE-KIT, reducing agent misbehaviors by 34% to 52% while simultaneously boosting task completion rates.

Zoom

Salesforce AI Research has introduced Prospective Hindsight (PH), a self-calibrating training framework for long-horizon autonomous agents. Addressing the fundamental limitation of standard RL where credit assignment occurs strictly after environmental feedback, PH exploits the gap between the agent's pre-action prospective prediction and post-action retrospective evaluation. By weighting gradients with this surprise residual, PH eliminates overconfident failures in single-turn tasks and underconfident successes in multi-turn interactions across GRPO and on-policy distillation.

Zoom

Reinforcement Learning with Verifiable Rewards (RLVR) powers modern reasoning LLMs by translating post-interaction scalar outcomes into optimization signals. However, in group-relative objectives (e.g., GRPO), reward signals vanish completely when all rollouts within a group share the same score (such as widespread mutual failures), discarding rich diagnostic feedback regarding task requirements and failure modes. In smaller models (such as 2B parameters), up to 98% of rollout groups result in universal failure, leaving standard RLVR stranded at 0% task success. Salesforce AI Research introduces Self-Retrospection Distillation (SRD, arXiv:2610.08077, repo: github.com/SalesforceAIResearch/SRD). SRD instantiates prospective learning, using privileged post-hoc trajectories to supervise pre-interaction foresight representations within the same policy without adding inference latency. Across 10 tool-integrated reasoning and long-horizon agentic benchmarks, SRD delivers gains of up to 24.2 percentage points over baseline RLVR and self-distillation. Critically, in 2B model regimes where 98% of groups are all-failure and standard RLVR flatlines at 0.0%, adding SRD reaches a 60.6% success rate under identical rollout budgets.

Zoom

On 2026-10-08 Anthropic launched Claude Dashboards (paid-plan beta; BigQuery, Snowflake, Databricks, Salesforce and more, with the SQL behind every chart) and Claude Motion (Team/Enterprise beta; code-built, editable animations with no video model, MP4 export). Docs, Slides, and Design left beta on all plans including Free, with 45M+ created so far.

Zoom

Claude (@claudeai) launched Claude Marketplace—one place for connectors/plugins, agents/products, and service partners. Customers get 2,000+ connectors (Atlassian, Google, Microsoft, Notion, Salesforce, …), can apply Anthropic committed spend to Claude-powered software from CrowdStrike, Cursor, Harvey, Legora, Lovable, Snowflake, and engage Accenture/BCG/Deloitte-style partners. Builders can list to reach Claude teams.

Zoom