Phone Segmentation and Recognition through Phonological Activation Mapping
Paper • 2607.09020 • Published • 9
AI & ML interests
voice-conversion speech-separation speech-enhancement speech-translation speech-synthesis speech-recognition spoken-language-understanding
Recent Activity
shikhar7ssu
authored a
paper 1 day ago
Add pipeline tag, library name, paper link, and usage instructions
#1 opened 8 days ago
by 
nielsr
Add pipeline tag, library name, and usage instructions
#1 opened 8 days ago
by nielsr
Add pipeline tag, library name, paper link, and usage section
#1 opened 8 days ago
by nielsr
shikhar7ssu
submitted a
paper to Daily Papers 9 days ago
RuiRuihigh
updated a
model 11 days ago
Post
285
Uncensoring Mistral,
give it a try
s3nh/Ministral-3-14B-Instruct-2512-BF16-abliterated
give it a try
s3nh/Ministral-3-14B-Instruct-2512-BF16-abliterated
RuiRuihigh
published a
model 12 days ago
JinchuanTian
updated a
dataset 15 days ago
JinchuanTian
published a
dataset 23 days ago
app.py
#4 opened about 1 month ago
by 
nathanjames
app.py
#5 opened about 1 month ago
by nathanjames
RishabA
updated a
model about 1 month ago
huckiyang
authored 2
papers about 1 month ago
RishabA
published a
model about 1 month ago
Post
352
Existing methods — GPTQ, AWQ, llama.cpp's k-quants — minimize empirical loss heuristically. None of them prove they are optimal in any information-theoretic sense. ICRB-Q builds a quantization scheme that is provably optimal via the Cramér-Rao lower bound (CRB): no unbiased estimator of a weight can have lower variance than [F(θ)]⁻¹, where F is the Fisher information matrix.
- 1 reply
vectominist
authored 2
papers about 1 month ago
d3bach
authored a
paper about 2 months ago