When the Right Answer Is No Answer: Teaching AI to Refuse Messy Math
MathDoc shows why document AI needs calibrated refusal, not just better transcription, when real exam papers are noisy, occluded, and incomplete.
Cognaptus Insights
Industry thoughts, trends, and commentary.
Explaining the Explainers: Why Faithful XAI for LLMs Finally Needs a Benchmark
A mechanism-first reading of LIBERTy, a structural-counterfactual benchmark that tests whether concept-based explanations actually track causal model behavior rather than merely producing plausible edits.
GUI-Eyes: When Agents Learn Where to Look
GUI-Eyes shows why GUI agents need learned active perception, not just bigger models staring harder at screenshots.
MatchTIR: Stop Paying Every Token the Same Salary
MatchTIR shows why multi-turn tool agents need fine-grained credit assignment, not just bigger models or louder final-answer rewards.
Recommendations With Receipts: When LLMs Have to Prove They Behaved
A mechanism-first look at PCN-Rec, a proof-carrying architecture that turns LLM recommenders from trusted decision-makers into auditable proposers.
Scaling Laws Without Power Laws: Why Bigger Models Still Win
A mechanism-first reading of why transformer scaling laws can survive even when the data itself has no power-law structure.
Survival by Swiss Cheese: Why AI Doom Is a Layered Failure, Not a Single Bet
A business-facing reading of AI existential risk as a portfolio of survival assumptions, not one melodramatic prediction.
When Memory Stops Guessing: Stitching Intent Back into Agent Memory
STITCH shows why long-horizon agents need memory indexed by task intent, not just larger context windows or better embeddings.
Bubble Trouble: Why Top‑K Retrieval Keeps Letting LLMs Down
A practical reading of Context Bubble construction: why enterprise RAG needs constrained, auditable context assembly rather than larger top-k piles.
Drawing with Ghost Hands: When GenAI Helps Architects — and When It Quietly Undermines Them
A mechanism-first reading of experimental evidence showing why GenAI helps novice architectural designers, fails to broadly lift performance, and can quietly weaken creative agency.