·
DataBubble
  • Home
  • Models
  • News
  • Compare
  • Boards
  • Pricing
  • About
  • Newsletter
  • Methodology
  • Contact
Latest
The agentic web is here4h◆Enhancing Sentiment Classification on Small Datasets through Data Augmentation and Transfer Learning8h◆Comparative Evaluation of LLM-Based Review Paradigms for Academic Paper Quality Assessment8h◆BlindToChange: AI Agents Can Find Changes When Asked But Do They Look When Needed?8h◆DAPS: Dependency-Aware Premise Selection for LLM Theorem Proving8h◆KV-Cache Patching: Tracing How and Where Fine-Tuned Behaviors Arise8h◆KEEPER: Learning Control-Sufficient Textual Memory for Long-Horizon LLM Agents8h◆Sparse Text Diffusion under Dual Budgets: Context Connectivity and Expert Activation8h◆Enhancing Knowledge Entity Relation Extraction in Abstracts and Full-text Content of Academic Articles Using Large Language Models8h◆CEFR Level Assessment across Languages and Text Granularities8h◆Digital sovereignty in professional translation workflows: a case study of migrating from Windows to Linux8h◆BYOD: Build Your Own DLM - Efficient Autoregressive-to-Diffusion Language Model Conversion8h◆Saidzoda-Tajik-LFM-2.6B: Adapting a 2.6B Hybrid Model to Tajik Tool Calling, and What Continued Pretraining Takes Away8h◆When Steering Shifts Discourse: Identifying and Intervening on Discourse-Level Changes due to Activation Steering in LLMs8h◆Precision--Recall Trade-off on the Inference Level for LLMs in Grammatical Error Correction8h◆Peeling the Matryoshka: Extracting Ontological Structure from Retrieval Embeddings8h◆Iterated Integrals as Lexical Fingerprints: Path Signature Embeddings for Lexical Similarity8h◆SimXAI: A Simple Drop-in Layer Increases Interaction Quality and Understanding in Conversational XAI8h◆Text as an Index for Zero-Shot Cross-Domain Knowledge Tracing8h◆Aligning MLLMs with the Latent Structure of Human Cognition via Behavior-Derived Semantic Dimensions8h◆The agentic web is here4h◆Enhancing Sentiment Classification on Small Datasets through Data Augmentation and Transfer Learning8h◆Comparative Evaluation of LLM-Based Review Paradigms for Academic Paper Quality Assessment8h◆BlindToChange: AI Agents Can Find Changes When Asked But Do They Look When Needed?8h◆DAPS: Dependency-Aware Premise Selection for LLM Theorem Proving8h◆KV-Cache Patching: Tracing How and Where Fine-Tuned Behaviors Arise8h◆KEEPER: Learning Control-Sufficient Textual Memory for Long-Horizon LLM Agents8h◆Sparse Text Diffusion under Dual Budgets: Context Connectivity and Expert Activation8h◆Enhancing Knowledge Entity Relation Extraction in Abstracts and Full-text Content of Academic Articles Using Large Language Models8h◆CEFR Level Assessment across Languages and Text Granularities8h◆Digital sovereignty in professional translation workflows: a case study of migrating from Windows to Linux8h◆BYOD: Build Your Own DLM - Efficient Autoregressive-to-Diffusion Language Model Conversion8h◆Saidzoda-Tajik-LFM-2.6B: Adapting a 2.6B Hybrid Model to Tajik Tool Calling, and What Continued Pretraining Takes Away8h◆When Steering Shifts Discourse: Identifying and Intervening on Discourse-Level Changes due to Activation Steering in LLMs8h◆Precision--Recall Trade-off on the Inference Level for LLMs in Grammatical Error Correction8h◆Peeling the Matryoshka: Extracting Ontological Structure from Retrieval Embeddings8h◆Iterated Integrals as Lexical Fingerprints: Path Signature Embeddings for Lexical Similarity8h◆SimXAI: A Simple Drop-in Layer Increases Interaction Quality and Understanding in Conversational XAI8h◆Text as an Index for Zero-Shot Cross-Domain Knowledge Tracing8h◆Aligning MLLMs with the Latent Structure of Human Cognition via Behavior-Derived Semantic Dimensions8h◆
News/KV-Cache Patching: Tracing How and Where Fine-Tuned Behaviors Arise
arxiv
PublishedNovember 2, 2026 at 5:00 AM

KV-Cache Patching: Tracing How and Where Fine-Tuned Behaviors Arise

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

arXiv:2610.41480v1 Announce Type: new Abstract: Fine-tuning a language model changes how it responds, but where is this fine-tuned behavior instantiated? To better understand this, we evaluate whether the change in model behavior arises only during the generation of the answer tokens where the chang

Stay posted· Newsletter

Three short emails a week: Monday, Wednesday and Friday. Top movers, new models and the most-referenced story.

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

Your email is used only to send The Bubble Brief. Privacy policy

Discussion
Source
↗
arxiv
Read original ↗All from arxiv →

No replies yet. Be first.

Source
↗
arxiv
Read original ↗All from arxiv →

Related coverage

More from ARXIV
arxivEnhancing Sentiment Classification on Small Datasets through Data Augmentation and Transfer Learning8harxivComparative Evaluation of LLM-Based Review Paradigms for Academic Paper Quality Assessment8harxivBlindToChange: AI Agents Can Find Changes When Asked But Do They Look When Needed?8harxivDAPS: Dependency-Aware Premise Selection for LLM Theorem Proving8h
The Bubble Brief
MON · WED · FRI

Read AI insights every Monday, Wednesday and Friday — top movers, new models, story of the week.

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

Your email is used only to send The Bubble Brief. Privacy policy

Originally published on arxiv ↗
PrivacyTermsRefundsLegal notice|AboutMethodologyContact

© 2026 Databubble

Built by Marouane Gazouzi · Readyo, AI interview coach

HomeModelsNews