{
  "schema_version": "1.1",
  "id": "archive:https://arxiv.org/abs/2610.35846v1",
  "slug": "2610-35846v1-0krwatc",
  "url": "https://feed7.dev/p/2610-35846v1-0krwatc",
  "title": "Preserve to Expand: Source-Aware Adaptation and Model Merging for Multilingual LLM-Based ASR",
  "why_included": "A study of multilingual speech-model adaptation finds a trade-off between learning new languages and preserving existing ones. Update merging and merge order can change how well the original languages are retained.",
  "summary": "The paper studies adapting a trainable speech connector while keeping the language model frozen. In single-target expansion, **SSU-Wanda+TIES** gave the best reported balance of learning a new language and retaining source-language performance; target-only tuning and standard TIES caused severe forgetting.",
  "practical_implication": "For speech-model builders adding languages without replaying source-language data, compare both target adaptation and source retention. In multi-target expansion, update merging retained source performance more consistently, while recursive expansion could improve target adaptation depending on language order.",
  "agent_context": "The paper studies adapting a trainable speech connector while keeping the language model frozen. In single-target expansion, **SSU-Wanda+TIES** gave the best reported balance of learning a new language and retaining source-language performance; target-only tuning and standard TIES caused severe forgetting.\n\nFor speech-model builders adding languages without replaying source-language data, compare both target adaptation and source retention. In multi-target expansion, update merging retained source performance more consistently, while recursive expansion could improve target adaptation depending on language order.\n\nThe findings concern connector-based Speech LLMs. The material does not provide enough detail to generalize the results to other model architectures or training setups.",
  "source": {
    "name": "arXiv",
    "url": "https://arxiv.org/abs/2610.35846v1",
    "published_at": "2026-10-26T17:56:22.000Z"
  },
  "source_class": "blog_post",
  "content_type": "Paper",
  "layer": "model",
  "domains": [
    "audio",
    "research"
  ],
  "topics": [
    "open-models",
    "model-selection"
  ],
  "verification": {
    "status": "needs_review",
    "label": "Needs Review",
    "method": "unverified",
    "verified_at": null
  },
  "uncertainty": [
    "The findings concern connector-based Speech LLMs. The material does not provide enough detail to generalize the results to other model architectures or training setups."
  ],
  "connected_context": {
    "meaning": "This supplies a concrete, connector-specific example of multilingual expansion where adaptation quality depends on retaining source-language ability as well as learning the target. The results favor particular merge strategies in the tested Speech LLM setup, so they sharpen adaptation choices without establishing a general recipe for other architectures.",
    "corpus_size": 928,
    "generated_at": "2026-10-27T11:06:56.591Z",
    "connections": [
      {
        "title": "Adaption Labs: Gradient-Free Continual Learning — Sara Hooker, Adaption",
        "source_name": "AI Engineer",
        "source_url": "https://www.youtube.com/watch?v=XEd_SRVHBgU",
        "feed7_url": "https://feed7.dev/p/adaption-labs-gradient-free-continual-learning-sara-hooker-adaption-1pozjjc",
        "reason": "Both concern adapting models while preserving useful existing capability; this paper reports specific connector-merging and retention outcomes, whereas Auto Scientist describes broader automated search over adaptation recipes without comparable evidence here."
      },
      {
        "title": "Program-as-Weights: A Programming Paradigm for Fuzzy Functions",
        "source_name": "arXiv",
        "source_url": "https://arxiv.org/abs/2607.02512v1",
        "feed7_url": "https://feed7.dev/p/2607-02512v1-1dr5458",
        "reason": "Both adapt a frozen language-model backbone through trainable added components, but the target studies speech connectors and language retention while Program-as-Weights addresses fuzzy-function execution."
      }
    ]
  },
  "lifecycle": "Current",
  "published_at": "2026-10-26T17:56:22.000Z",
  "modified_at": "2026-10-26T17:56:22.000Z",
  "supersedes": [],
  "expires_at": null,
  "formats": {
    "html": "https://feed7.dev/p/2610-35846v1-0krwatc",
    "json": "https://feed7.dev/p/2610-35846v1-0krwatc.json",
    "markdown": "https://feed7.dev/p/2610-35846v1-0krwatc.md"
  }
}