·
DataBubble
  • Home
  • Models
  • News
  • Compare
  • Boards
  • Pricing
  • About
  • Newsletter
  • Methodology
  • Contact
Latest
Google is working on a new AI chip designed to make Gemini more efficient1h◆SpaceX in your index fund, explained1h◆AI’s most important protocol is getting a little bit easier to use2h◆X relaunches a rebuilt Android app after year-long effort3h◆OpenAI is scared of open-weight models. Should the US be?3h◆China’s AI models have Trump’s AI world at war with itself4h◆Adobe’s ‘natural look’ camera app embraces generative AI6h◆Introducing Cosmos 3 Edge6h◆Adobe camera app’s new feature will critique your photos using AI7h◆YouTube clarifies policies around AI slop and upsetting videos7h◆China delivers a one-two punch to America’s AI dominance12h◆Safety and alignment in an era of long-horizon models12h◆AI is more likely than humans to form biases when hiring14h◆ADS-C: Antidistillation Sampling for Classification18h◆Beyond a Joke: Multi-Angle Reasoning for Detecting and Explaining Harmful Humor in Memes18h◆From Black Box to Executable Logic: Explainable Reinforcement Learning through Prolog Expert Systems18h◆A Formally Grounded ODRL Evaluator: Implementation and Comparison18h◆Efficient Difficulty-Aware Dynamic Routing for Diffusion-Based Real-World Image Super-Resolution18h◆Map as a Prompt: Learning Multi-Modal Spatial-Signal Foundation Models for Cross-scenario Wireless Localization18h◆The AI Fiction Paradox18h◆Google is working on a new AI chip designed to make Gemini more efficient1h◆SpaceX in your index fund, explained1h◆AI’s most important protocol is getting a little bit easier to use2h◆X relaunches a rebuilt Android app after year-long effort3h◆OpenAI is scared of open-weight models. Should the US be?3h◆China’s AI models have Trump’s AI world at war with itself4h◆Adobe’s ‘natural look’ camera app embraces generative AI6h◆Introducing Cosmos 3 Edge6h◆Adobe camera app’s new feature will critique your photos using AI7h◆YouTube clarifies policies around AI slop and upsetting videos7h◆China delivers a one-two punch to America’s AI dominance12h◆Safety and alignment in an era of long-horizon models12h◆AI is more likely than humans to form biases when hiring14h◆ADS-C: Antidistillation Sampling for Classification18h◆Beyond a Joke: Multi-Angle Reasoning for Detecting and Explaining Harmful Humor in Memes18h◆From Black Box to Executable Logic: Explainable Reinforcement Learning through Prolog Expert Systems18h◆A Formally Grounded ODRL Evaluator: Implementation and Comparison18h◆Efficient Difficulty-Aware Dynamic Routing for Diffusion-Based Real-World Image Super-Resolution18h◆Map as a Prompt: Learning Multi-Modal Spatial-Signal Foundation Models for Cross-scenario Wireless Localization18h◆The AI Fiction Paradox18h◆
News/Countering Catastrophic Forgetting of Large Language Models for Better Instruction Following via Weight-Space Model Merging
arxiv
PublishedApril 4, 2026 at 4:00 AM
▲bullish

Countering Catastrophic Forgetting of Large Language Models for Better Instruction Following via Weight-Space Model Merging

Source
arxiv.orgfull article ↗
Read on arxiv→
Publisher summary· verbatim

arXiv:2604.01538v1 Announce Type: new Abstract: Large language models have been adopted in the medical domain for clinical documentation to reduce clinician burden. However, studies have reported that LLMs often "forget" a significant amount of instruction-following ability when fine-tuned using a t

Models mentioned
01
  • 01meta-llama logo
    Llama-3.1-8B
    meta-llama/Llama-3.1-8B
    DL 1.5M0.0%IN $0.10/Mtok
Related
04
  • arxivMay 28
    A Benchmark Construction and Evaluation Framework for Specialist Domains: Case Study on Defense-related Documents
  • arxivMay 22
    GraphRAG on Consumer Hardware: Benchmarking Local LLMs for Healthcare EHR Schema Retrieval
  • arxivMay 22
    DrugRAG: Enhancing Pharmacy LLM Performance Through A Novel Retrieval-Augmented Generation Pipeline
  • arxivMay 8
    Measuring Evaluation-Context Divergence in Open-Weight LLMs: A Paired-Prompt Protocol with Pilot Evidence of Alignment-Pipeline-Specific Heterogeneity
Stay posted· Newsletter

A 5-min weekly brief — top movers, price watch, story of the week.

// no spam · unsubscribe one-click · free forever

Discussion
Mentioned models
02
  • 01
    GatorTronLlama
  • 02
    Llama-3.1-8B
    meta-llama/Llama-3.1-8B
    1.5M dl
Source
↗
arxiv
Read original ↗All from arxiv →
Tags
04
#open-source#clinical#domain-adaptation#language-models

No replies yet. Be first.

Mentioned models
02
  • 01
    GatorTronLlama
  • 02
    Llama-3.1-8B
    meta-llama/Llama-3.1-8B
    1.5M dl
Source
↗
arxiv
Read original ↗All from arxiv →
Tags
04
#open-source#clinical#domain-adaptation#language-models

Related coverage

More from ARXIV
arxivADS-C: Antidistillation Sampling for Classification18harxivBeyond a Joke: Multi-Angle Reasoning for Detecting and Explaining Harmful Humor in Memes18harxivFrom Black Box to Executable Logic: Explainable Reinforcement Learning through Prolog Expert Systems18harxivA Formally Grounded ODRL Evaluator: Implementation and Comparison18h
The Bubble Brief
WEEKLY

Read open-source insights every Tuesday — top movers, new releases, story of the week.

// no spam · unsubscribe one-click · free forever

Originally published on arxiv ↗
HomeModelsNews