🤖 ai - fyenne

🧠AI Research Academic

arxiv.org·

Instruction Finetuning DeepSeek-R1-8B Model Using LoRA and NEFTune

⚙️ML Engineering Academic

arxiv.org·

Distilling Safe LLM Systems via Soft Prompts for On Device Settings

⚙️ML Engineering Academic

arxiv.org·

Advancing the State-of-the-Art in Empirical Privacy Auditing

⚙️ML Engineering Academic

arxiv.org·

When RL Fails after SFT: Rejuvenating Model Plasticity for Robust SFT-to-RL Handoff

⚙️ML Engineering Academic

arxiv.org·

EstRTL: Functional Estimation Guided RTL Code Generation

⚙️ML Engineering Academic

arxiv.org·

MemToolAgent overview with a simple restaurant booking scenario where the agent retrieves similar memories, receives feedback on an invalid time format, and generates a reflection to update its memory

⚙️ML Engineering Academic

arxiv.org·

New comment by Ayaz_Saifi in "Ask HN: Who wants to be hired? (June 2026)"

⚙️ML Engineering

drive.google.com··Hacker News

Alignment Defends LLMs from Property Inference Attacks

💬LLMs Academic

arxiv.org·

harshuljain13/llm-inference-at-scale: A Practitioner handbook for production llm serving.

⚙️ML Engineering Code

github.com··Hacker News

Uncertainty-Aware LLM-Guided Policy Shaping for Sparse-Reward Reinforcement Learning

⚙️ML Engineering Academic

arxiv.org·

Pushing the Limits of LLM Tool Calling via Experiential Knowledge Integration and Activation

⚙️ML Engineering Academic

arxiv.org·

The Shibboleth Effect: Auditing the Cross-Lingual Distributional Skew of Large Language Models

💬LLMs Academic

arxiv.org·

What Are AI Hallucinations? And How To Improve Accuracy in Pipelines

My research agenda and work

Fisher-Guided Progressive Parameter Selection for Adaptive Fine-Tuning

The Transformer, Demystified — Let's Actually Build One

Mycol: A user-friendly app for automating analysis of microscopy images

Evaluating Hallucinations in Domain-Adapted Large Language Models

Efficient and Training-Free Single-Image Diffusion Models

Machine Learning Methods for Studying Latent Neural Activity Dynamics

Instruction Finetuning DeepSeek-R1-8B Model Using LoRA and NEFTune

Distilling Safe LLM Systems via Soft Prompts for On Device Settings

Advancing the State-of-the-Art in Empirical Privacy Auditing

When RL Fails after SFT: Rejuvenating Model Plasticity for Robust SFT-to-RL Handoff

EstRTL: Functional Estimation Guided RTL Code Generation

MemToolAgent overview with a simple restaurant booking scenario where the agent retrieves similar memories, receives feedback on an invalid time format, and generates a reflection to update its memory

New comment by Ayaz_Saifi in "Ask HN: Who wants to be hired? (June 2026)"

Alignment Defends LLMs from Property Inference Attacks

harshuljain13/llm-inference-at-scale: A Practitioner handbook for production llm serving.

Uncertainty-Aware LLM-Guided Policy Shaping for Sparse-Reward Reinforcement Learning

Pushing the Limits of LLM Tool Calling via Experiential Knowledge Integration and Activation

The Shibboleth Effect: Auditing the Cross-Lingual Distributional Skew of Large Language Models