8). Mitigating LLM Jailbreaks with Few Examples – introduces a new approach called for defending LLMs against jailbreak attacks, focusing on quickly adapting defenses after detecting new attacks rather than aiming for perfect adversarial upfront robustness.
RESEARCH
-
Test-Time Training Boosts LLM Abstract Reasoning on ARC
By
–
5). The Surprising Effectiveness of Test-Time Training for Abstract Reasoning – explores test-time training (TTT) – updating model parameters temporarily during inference – for improving an LLM's abstract reasoning capabilities using the ARC benchmark.
-
OpenCoder: Open-Source LLM for Code Generation
By
–
4). OpenCoder – introduces OpenCoder, a fully open-source LLM specialized for code generation and understanding; the authors identify several critical factors for building high-performing code LLMs: (1) effective data cleaning with code-optimized heuristic rules for
-
AgentOps Taxonomy: Observability for Foundation Model-Based Agents
By
–
6). A Taxonomy of AgentOps for Enabling Observability of Foundation Model-based Agents – analyzes AgentOps platforms and tools, highlighting the need for comprehensive observability and traceability features to ensure reliability in foundation model-based autonomous agent systems
-
Precision-Aware Scaling Laws for Large Language Models
By
–
2). Scaling Laws for Precision – introduces "precision-aware" scaling laws that predict how model performance is affected by both training and inference precision in LLMs.
-
Evo: 7B DNA Sequence AI Model Processes 131 Kilobases
By
–
3). Evo – a 7B parameter AI model designed to understand and generate DNA sequences across multiple biological scales; the model, trained on 2.7 million prokaryotic and phage genomes, can process sequences up to 131 kilobases long while maintaining single-nucleotide resolution…
-
Top ML Papers of the Week: RAG, Transformers, and Optimization
By
–
The Top ML Papers of the Week (Nov 11 – 17): – HtmlRAG
– OpenCoder
– Mixture of Transformer
– A Taxonomy of AgentOps
– Scaling Laws for Precision
– Toward Optimal Search and Retrieval for RAG Read on for more: -
How Top Scientists Leverage AI for Innovation Success
By
–
1). Impacts of AI on Innovation – suggests that top scientists leverage their domain knowledge to prioritize promising AI suggestions, while others waste significant resources testing false positives…
-
Diversity of AI Hardware Accelerators and Software Stacks
By
–
There is a wide diversity of specialized HW accelerators and SW stacks designed and built to run AI applications efficiently, so it is challenging to understand their differences, capabilities, programming approaches, and how they perform, particularly for scientific
-
Dynamic Section Retrieval Demo Notebook LlamaParse
By
–
Dynamic section retrieval, demo notebook: https://
github.com/run-llama/llam
a_parse/blob/main/examples/advanced_rag/dynamic_section_retrieval.ipynb
…