University of California, Berkeley

MMLU was released by the University of California, Berkeley in 2020 and contains approximately 15,000 multiple-choice questions spanning 57 subjects.

BFCL (Berkeley Function Calling Leaderboard) is a tool-calling capability evaluation benchmark released by the University of California, Berkeley, with V4 being its multi-turn dialogue extension.

The vLLM project was initiated by the UC Berkeley team.

LM Arena (formerly Chatbot Arena) is an anonymous head-to-head large model evaluation platform developed by a team at the University of California, Berkeley.

RISC-V is an open standard instruction set architecture based on reduced instruction set principles, initiated by the University of California, Berkeley in 2010, with the instruction set specification fully open-source and free to use.

LMARENA (formerly LMSYS Chatbot Arena) is a large model capability evaluation platform initiated by the University of California, Berkeley, using a human blind-test battle mechanism.

vLLM由加州大学伯克利分校开发，其核心创新PagedAttention技术相比原生HuggingFace Transformers推理可实现数倍到数十倍的吞吐量提升

吴恩达在卡内基梅隆大学获得计算机科学学士学位，在MIT获得硕士学位，在加州大学伯克利分校获得博士学位

Researchers from Stanford University and UC Berkeley proposed a method called TokenGate that accelerates long-context prefill through a Token-Gated Attention mechanism.

Tīrtha Bench's core test contains 800 tool-calling tasks, taken from Berkeley's publicly released BFCL dataset, of which 400 are single calls, 200 are multi-tool selection, and 200 are tasks requiring parallel calls.

LMArena (formerly Chatbot Arena) is a research project from UC Berkeley and other institutions, originally used to evaluate the relative capabilities of large language models in a crowdsourced way.

When designing the 8087, Intel collaborated with Professor William Kahan of UC Berkeley to establish a floating-point model centered on the hidden bit, normalized numbers, four rounding modes, and five exception classes.

FrameNet 是加州大学伯克利分校于1997年启动的计算语言学资源，覆盖上千个语义框架与数十万条标注例句。

The 1973 UC Berkeley admissions data appeared overall to show a gender bias against women, but when broken down by department, most departments had higher admission rates for women; the disparity stemmed from women tending to apply to more competitive departments.

BFCL is a specialized benchmark released by the University of California, Berkeley, that evaluates the accuracy of large language models in structured function-calling scenarios.

Stanford University and the University of California, Berkeley are the talent sources of the Bay Area's robotics ecosystem.

RISC-V was released by the University of California, Berkeley around 2010, and is an open instruction set architecture (ISA) specification rather than a complete chip implementation.

All seven co-founders of Databricks are graduates of the University of California, Berkeley.

SPICE originally originated at the University of California, Berkeley.

vLLM was originally proposed by a research team at the University of California, Berkeley.