For Kocree, explainability is more than a reporting feature. It is a way to make scientific knowledge actionable—and turn a company’s accumulated research into an asset that grows exponentially more valuable with use.
That is a commercial opportunity: turn a company’s proprietary R&D data into design knowledge (with full IP traceability, protection, and ownership). The goal is faster iteration, less wasted experimentation, and more value from every R&D dollar.
He also identifies the capability that matters most for discovery: “The lattice becomes a map of informational dependencies—capturing not only how knowledge is distributed, but also how it can be recombined to form new insights.”
Rajan writes: “It enables the identification of patterns that persist across scales and the discovery of rules that bridge microscopic descriptors with macroscopic responses. By mapping how informational dependencies propagate through this hierarchy, ILL not only enhances explainability but also offers a formal mechanism for understanding—for uncovering the principles that link structure, processing, and property in complex material systems.”
When design constraints change, teams can revisit the underlying rules, examine which still apply, and investigate new combinations—rather than treating every project as a fresh prediction problem.
In R&D workflows, our ILL technology can guide substitutions for scarce elements, explore new chemical–structural combinations, and focus expensive experiments on better-motivated candidates. The output is usable knowledge, not just another prediction score.
The underlying ILL technology is receiving substantive expert recognition. Krishna Rajan—Humboldt Prize winner and internationally-recognized pioneer of materials informatics—gives ILL an extended treatment in his new book, Materials Informatics: Molecules, Crystals, and Beyond. (Figure 1.15 from the book)
Kocree is building AI that unlocks a company’s R&D knowledge: learn the rules behind existing results, combine them under new constraints, and create candidates that R&D teams can test.
🧵[1/N] Are large language models more creative than humans at combining ideas? 🤔
Combinatorial creativity involves combining familiar ideas in unfamiliar ways, and is central to scientific discovery and technological innovation.
In fact, breakthrough scientific work is often associated with more distant combinations of methods, techniques, and concepts.
Here, we introduce Kombine, a new benchmark of combinatorial creativity for LLMs.
TLDR:
- LLMs are more combinatorially creative than the average human.
- However, conceptual blending is the hardest task for LLMs, and even frontier models struggle.
- LLMs also tend to homogenize, inventing concepts with the same properties across separate models.
- Moreover, humans and LLMs combine in distinct regions of semantic space.
- As a result, we anticipate greater conceptual diversity can be unlocked through human-AI collaboration. 🗣️
Thread below🧵⬇️
How much do LLMs’ watermarks distort your output? 🔍
We present SimpleMark: a fast multi-bit text watermark that bounds the distortion of every token to a chosen f-divergence budget. Our watermark’s text distribution stays statistically close to what the model would have written without a watermark while remaining competitive in all benchmarked performance categories.
Full transparency, traceability, attribution, and therefore credit is what we're building at Kocree! Drawing on our novel information lattice learning AI, integrated with trust technologies. Scientific creativity, enterprise creativity, and artistic creativity.
i once asked a bunch of deep-tech researchers on why they wouldn’t use chatgpt for serious research, especially after openai launched Prism tool around a workflow researchers already know well: LaTeX. Their responses were surprisingly consistent: they were afraid that putting genuinely novel, unpublished ideas into chatgpt or any other llms could expose or compromise their research originality.
Whether or not that fear is technically justified, the perception itself is a problem. If OpenAI wants researchers to trust these tools with high-stakes research, there needs to be much more explicit transparency around who can access conversations, how research data is retained, whether it is used for training, and what protections exist for unpublished IP.
i assume frontier labs can access at least ~70% of user conversation details internally, even if PII is stripped or access is tightly controlled?
When I started Canonical, the mission was simple: back technical founders building core mission-critical infrastructure before it's obvious.
Today, I'm extremely proud to share that Canonical has been selected for the 2026 @raiseconference Select cohort alongside an incredible group of emerging fund managers.
To our portfolio founders, LPs, and community partners - thank you for trusting us.
Looking forward to connecting with LPs & GPs at The Presidio in SF on Oct 22, 2026. Drop a DM if you'll be there!
The LLMs’ inherent inability to trace back their thought process and give credit where it’s due will only exacerbate the problem. (E.g., OpenAI’s latest dump of 10 proofs suffers from this problem.) 32/