MLSP 2026 keynote highlights trustworthy AI, world discovery and long-horizon memory
mohitban47 · x · 2026-10-05
At the 36th IEEE MLSP 2026 workshop in Atlanta, the author gave a keynote on key AI challenges: trustworthy and safe collaboration (including adversarial scenarios), skill+world discovery, and long-horizon memory planning, with applications spanning early medical diagnosis and classroom engagement.
More from Research
- ASCENT: online test-time training lets LLM agents self-distill verified experience into weights — Haodong Lu · 2026-10-06
- PluginRSI: recursive harness improvement via reusable plugins beats whole-program search — Yaorui Shi · 2026-10-06
- NVIDIA's OSWorld-Pro adds 2,800+ subgoal process evaluation — Claude Opus 5 drops to 75.7% — nvidia · 2026-10-06
- Tencent Hunyuan's Prism: dynamic sparse attention speeds up 2K joint video-audio training 2.5x — Tencent-Hunyuan · 2026-10-06
- PaLoRA derives rank-aware pacing law for LoRA continual learning, +4% on 50-task benchmarks — Yuxuan Li · 2026-10-06
- LoHi: dense low-res sampling lifts long-video VLM accuracy 10.6 pts and cuts decoding latency 7x — jchoi2012 · 2026-10-06