pith:R4Y3VJSP
The Silent Thought: Modeling Internal Cognition in Full-Duplex Spoken Dialogue Models via Latent Reasoning
Spoken dialogue models can perform continuous internal reasoning while listening by recursively updating latent embeddings.
arxiv:2603.17837 v4 · 2026-03-18 · eess.AS · cs.CL
Add to your LaTeX paper
\usepackage{pith}
\pithnumber{R4Y3VJSPSUGJCEGU7RYYSFULMP}
Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge
Record completeness
Claims
our approach aligns seamlessly with spoken dialogue systems: during the user's speaking phase, it recursively feeds the latent embedding output from the previous step into the next step, enabling continuous reasoning that strictly adheres to causality without introducing additional latency.
That recursive updates to latent embeddings can meaningfully represent and advance internal cognitive processing without explicit reasoning supervision or causing drift in the dialogue state.
FLAIR enables spoken dialogue AI to conduct continuous latent reasoning while perceiving speech through recursive latent embeddings and an ELBO-based finetuning objective.
Cited by
Receipt and verification
| First computed | 2026-05-21T01:05:17.921259Z |
|---|---|
| Builder | pith-number-builder-2026-05-17-v1 |
| Signature | Pith Ed25519
(pith-v1-2026-05) · public key |
| Schema | pith-number/v1.0 |
Canonical hash
8f31baa64f950c9110d4fc7189168b63e66a86c17eabbf2113dad486770bb187
Aliases
· · · · ·Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/R4Y3VJSPSUGJCEGU7RYYSFULMP \
| jq -c '.canonical_record' \
| python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: 8f31baa64f950c9110d4fc7189168b63e66a86c17eabbf2113dad486770bb187
Canonical record JSON
{
"metadata": {
"abstract_canon_sha256": "7ef5b6f0546ee634d454410ff9aa6a4ad13752e6f8c9bc401bdd187fa84d03dd",
"cross_cats_sorted": [
"cs.CL"
],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"primary_cat": "eess.AS",
"submitted_at": "2026-03-18T15:30:29Z",
"title_canon_sha256": "d6d9110a3802ae95800846fa36cda7b2145e2fe02437d86fca1a182077ac8d08"
},
"schema_version": "1.0",
"source": {
"id": "2603.17837",
"kind": "arxiv",
"version": 4
}
}