ARC-Easy
PulseAugur coverage of ARC-Easy — every cluster mentioning ARC-Easy across labs, papers, and developer communities, ranked by signal.
2 day(s) with sentiment data
-
Quantized LLM Copies May Signal Model Errors
Researchers have investigated a novel method for detecting errors in small language models by comparing the outputs of two quantized versions of the same model. This approach, which involves an 8-bit copy answering a pr…
-
Runcrate automates nightly LLM evaluations on ephemeral GPUs
A new tool called Runcrate has been introduced to automate the nightly execution of LLM evaluations, ensuring that model regressions are detected promptly before releases. This system utilizes ephemeral GPUs that are pr…
-
New Arkios language model trained on English-Nepali text
Researchers have introduced Arkios, a 1.04 billion parameter language model trained on 150 billion tokens of English and Nepali text. The model utilizes a custom training stack and a Devanagari-aware tokenizer. Evaluati…
-
New Daedalus-150M model achieves faster CPU inference with hybrid architecture
Researchers have developed Daedalus-150M, a novel language model optimized for efficient CPU inference. This hybrid model combines sparse attention with short convolutions, allowing two-thirds of its architecture to avo…
-
New research explores LLM efficiency and reasoning improvements
Several research papers explore methods to enhance the efficiency and reliability of large language models (LLMs). Hugging Face's LFM2.5-DSpark demonstrates up to 3.2x faster inference speeds by using speculative decodi…
-
Constitutional Midtraining Enhances AI Alignment Durability
Researchers have developed a method called constitutional midtraining to improve the durability of AI alignment. By integrating principled, values-based content into the midtraining phase of AI development, models demon…
-
Constitutional Midtraining boosts AI alignment durability
Researchers have introduced "Constitutional Midtraining," a novel approach to enhance the durability of AI alignment. By embedding values-based content into the model's training phase, rather than solely after, this met…
-
Prism Transformer introduces progressive head schedules for hierarchical attention
Researchers have introduced the Prism Transformer, a novel architecture that modifies the standard multi-head attention mechanism. Instead of allocating equal dimensional space to each attention head at every layer, Pri…