work trail

Chronological record of public research entries.

May 10, 2026 essay Interestingness Is the Derivative of Beauty
A short formal note on Schmidhuber's compression-progress view: beauty as subjective description length, interestingness as its first derivative.
April 25, 2026 bioacoustics Udani: Depth-Recurrent Hybrids for Bioacoustic Representation Learning
Udani is a bioacoustic representation-learning codebase built around masked acoustic prediction, fused Gated DeltaNet blocks, local attention, and a depth-recurrent 3:1 hybrid backbone.
April 8, 2026 olfaction Topological Signals in Learned Odor Embeddings
A representation audit of OpenPOM and chemical baselines: robust H1 signal appears in learned odor embeddings, but the topology is not unique to POM.
March 9, 2026 speech Completed 12k Run: Early Speech Emergence in a 21.9M OLMo-Hybrid Speech LM
The full 12k-step A100 run of a 21.9M OLMo-hybrid speech codec LM on LJ Speech finished cleanly, reaching EMA val loss 3.8207 and perplexity 45.63, with clearly stronger samples than the earlier...
March 3, 2026 music MC-LA vs Hybrid MC-LA 1:3: Live Training (In Progress) in progress
Two memory-cached linear attention variants training on 25k FMA tracks. Early results: the hybrid nearly matches full MC-LA on loss while running 1.9x faster. GRM gating is learning selectivity...
March 1, 2026 music Memory Caching for SSMs: From Paper to Implementation
Applying Memory Caching (MC) from arXiv:2602.24281 to Mamba and linear attention for music generation — a novel extension the original paper never tested. Architecture decisions, Output Activation...
February 28, 2026 music SSMs for Music Generation: Baseline Experiments
Baseline comparison of Transformer vs Hybrid Mamba-Attention (1:3) for autoregressive music generation over DAC tokens. The hybrid matches the transformer on all codebook metrics while being...