IBBench-Light: A Paired Evaluation of Task-Conditioned Responses to External Directives
An external record may contain a procedure to apply or text to read, depending on the user's request. IBBench-Light tests both uses against the same record. Twelve semantic bases yield 144 matched pairs per model; four quantized instruction models produced 1,152 archived greedy responses. Paired exact-contract accuracy...
Kai-Nan Zhou, Zhao-Yi Li, Janet Sung et al.
· 0 citations