影响力指数
论文质量、代表作、近期表现、广度与样本量置信度综合计算
16.2/100
超过 49.1%
全站排名 #32,787
发表论文5 篇
平均评分
年均产出5.0 篇/年
Mubashara Akhtar
研究方向
multimodality · evaluation · benchmarking · datasets · AI Safety · responsible ai · Automated fact checking · fake news detection · Semi-structured data · tabular data
13
LEXam: Benchmarking Legal Reasoning on 340 Law Exams
ICLR 2026Poster
14
Compose and Fuse: Revisiting the Foundational Bottlenecks in Multimodal Reasoning
ICLR 2026Poster
9
Who Measures What Matters? An Analysis of Social Impact Evaluations in Foundation Model Reporting
ICLR 2026Desk Rejected
12
Reasoning with Confidence: Efficient Verification of LLM Reasoning Steps via Uncertainty Heads
ICLR 2026Withdrawn
16
Chimera: Diagnosing Shortcut Learning in Visual-Language Understanding
ICLR 2026Withdrawn