The surge in scientific submissions has placed increasing strain on the traditional peer-review process, prompting the exploration of large language models (LLMs) for automated review generation. While LLMs demonstrate competence in producing structured and coherent feedback, their capacity for critical reasoning, contextual grounding, and quality sensitivity remains limited. To systematically evaluate these aspects, we propose a comprehensive evaluation framework that integrates semantic similarity analysis and structured knowledge graph metrics to assess LLM-generated reviews against human-written counterparts. We construct a large-scale benchmark of 1,683 papers and 6,495 expert reviews from ICLR and NeurIPS in multiple years, and generate reviews using five LLMs. Our findings show that LLMs perform well in descriptive and affirmational content, capturing the main contributions and methodologies of the original work, with GPT-4o highlighted as an illustrative example, generating 15.74% more entities than human reviewers in the strengths section of good papers in ICLR 2025. However, they consistently underperform in identifying weaknesses, raising substantive questions, and adjusting feedback based on paper quality. GPT-4o produces 59.42% fewer entities than real reviewers in the weaknesses and increases node count by only 5.7% from good to weak papers, compared to 50% in human reviews. Similar trends are observed across all conferences, years, and models, providing empirical foundations for understanding the merits and defects of LLM-generated reviews and informing the development of future LLM-assisted reviewing tools. Data, code, and more detailed results are publicly available at https://github.com/RichardLRC/Peer-Review.

Deep joint source-channel coding (JSCC) with hybrid automatic repeat request (HARQ) for image transmission has attracted increasing attention due to its flexibility and high efficiency. Existing researches mainly focus on minimizing the distortion of mutiple retransmissions while ignoring the redundancy in retransmitted signal and such redundancy may lead bandwidth waste and reconstruction quality degradation. In this paper, we propose an information bottleneck (IB) guided deep JSCC with HARQ system (HARQ-IBJSC), which aims at improving the reconstruction quality by compressing the redundancy in retransmitted signal. In particular, we first design a new IB objective for deep JSCC with HARQ system, which simultaneously reduces redundancy in the retransmitted signal and minimizes image transmission distortion. Since the mutual information terms in the designed IB objective is intractable, we then derive a differentiable lower bound on the IB objective and use the bound as the loss function of HARQ-IBJSC. Experimental results show that the proposed HARQ-IBJSC system can increase PSNR by up to 1 dB.