Showing cs.LGShow all
3 papers · 1 filter
cs.LG2026
When Flat Minima Fail: Characterizing INT4 Quantization Collapse After FP32 Convergence
Marcus Armstrong
Post-training quantization (PTQ) assumes that a well-converged model is a quantization-ready model. We show this assumption fails in a structured, measurable, and previously unchar…
cs.LG2026
Dead Weights, Live Signals: Feedforward Graphs of Frozen Language Models
Marcus Armstrong, Navid Ayoobi, Arjun Mukherjee
We present a feedforward graph architecture in which heterogeneous frozen large language models serve as computational nodes, communicating through a shared continuous latent space…
cs.LG2026
Thinking in Different Spaces: Domain-Specific Latent Geometry Survives Cross-Architecture Translation
Marcus Armstrong, Navid Ayoobi, Arjun Mukherjee
We investigate whether independently trained language models converge to geometrically compatible latent representations, and whether this compatibility can be exploited to correct…