@mert_gulsun "MIght be" is a crazy was to say: "Definitely is"
MoE -> "Bad" for Local - since you need cap
Dense -> Better for Local, since every param is worth it
Loopy Boy (either/or) huge since TTS is way cheaper than cap, but Dense Loopy Boy is like Numero Uno
I think I’ve figured out the gameplay loop!
It’s a sports anime game, but the sport is elemental bending.
You fight in tournaments alongside your squad, and between matches they have their own stories you can get into, like in Mass Effect or Dragon Age.
Inkling 2 Mini is a looped transformer that runs on a laptop.
Looping lets a model think more and write less. At this scale, its thought process remains monitorable; this may not stay the case. We’re also sharing a variable-loop research model to enable further study.
> A common idea floating around is that LLMs will absorb harnesses and make them obsolete.
There is no way enough people believe this for it to be "common".
Harnesses used to be a crutch for weak LLMs. In 2026 and onward, with stronger models, harnesses now offer a mechanism for composing an LLM’s existing and new capabilities to generalize well beyond their training environments.
In our new paper, LM Harnesses are Compositional Generalizers, we explore how various harness abstractions enable LM systems to generalize over task length, task strategies, recursive task depth, and more through composing existing LLM capabilities. These abstractions provide a means for more data-efficient training, especially over long tasks, by keeping trajectories locally in-distribution (LID) — i.e., shaping all individual LLM calls to be in-distribution.
For tasks that take a very long time to roll out, this implies that we can even perform surrogate harness training, meaning we train an individual LLM call instead of doing the very long harness rollout, using purely synthetic tasks, with fake tool calls + sub-agents!
A common idea floating around is that LLMs will absorb harnesses and make them obsolete. We disagree, and argue that we haven’t sufficiently explored what harnesses can provide.
The full paper, logs, codebase, and data are available: https://t.co/AXBjdoqmiz
w/ @SOURADIPCHAKR18@lateinteraction
Got to have an amazing conversation with one of Germany's best and brightest!
@DadaJudith has written a lot, and participated a lot in the EU AI scene, and it was a blast to pick her brain at a local pub in Berlin!
AI is more than a better way to search.
Over a drink with NVIDIA’s Chris Alexiuk, @langdock Co-CEO @DadaJudith shared why AI’s biggest impact may be felt in better services, better products, and healthier, more affordable lives.
Watch their full conversation from GTC Berlin: https://t.co/0PLlhRXeWN