Papers
Abstract: This report synthesises findings from 13 peer-reviewed papers addressing the following research question: What is the effect of continuous user feedback integration on natural language understanding accuracy metrics in open-environment dialogue systems. 9 claims were extracted from source literature; 9 were…
Abstract: This report synthesises findings from 13 peer-reviewed papers addressing the following research question: How does the thinking mode in Qwen3 impact accuracy on GPQA Diamond compared to non-thinking modes in other frontier models. 8 claims were extracted from source literature; 8 were independently verified against…
Abstract: This report synthesises findings from 5 peer-reviewed papers addressing the following research question: How does on-the-job learning impact conversational coherence scores on the ConvEval benchmark compared to static pre-trained dialogue systems. 9 claims were extracted from source literature; 9 were independently…
Abstract: This report synthesises findings from 13 peer-reviewed papers addressing the following research question: How does chain-of-thought prompting impact accuracy on the MultiMedQA benchmark compared to zero-shot baselines across different model scales. 9 claims were extracted from source literature; 9 were independently…
Abstract: This report synthesises findings from 13 peer-reviewed papers addressing the following research question: How does the accuracy of multimodal LLMs on visual reasoning tasks (e.g., VQA v2, COCO-Caption) compare to that of text-only LLMs when given image descriptions as textual input. 7 claims were extracted from…
Abstract: This report synthesises findings from 13 peer-reviewed papers addressing the following research question: How does quantization-aware training affect multimodal alignment performance on the MME benchmark relative to post-training quantization methods. 16 claims were extracted from source literature; 13 were…
Abstract: This report synthesises findings from 11 peer-reviewed papers addressing the following research question: What is the degradation rate in HumanEval pass@1 scores for code generation tasks when applying dynamic quantization to transformer attention layers. 13 claims were extracted from source literature; 13 were…
Abstract: This report synthesises findings from 9 peer-reviewed papers addressing the following research question: What is the efficiency-performance tradeoff of MetaSC's dynamic safety specification optimization compared to fine-tuning proprietary models on safety benchmarks like BBH or MMLU. 15 claims were extracted from…
Abstract: This report synthesises findings from 16 peer-reviewed papers addressing the following research question: How does the performance of LVLMs scale with increasing model size when evaluated on LVLM-eHub's cross-domain tasks, and what is the optimal model size for balanced accuracy and efficiency. 10 claims were…
Abstract: This report synthesises findings from 14 peer-reviewed papers addressing the following research question: What is the comparative robustness of LVLMs in LVLM-eHub against adversarial attacks or noisy inputs, measured by accuracy degradation across different perturbation types. 12 claims were extracted from source…
Abstract: This report synthesises findings from 12 peer-reviewed papers addressing the following research question: What is the comparative impact of emotional intelligence alignment techniques on conversational coherence metrics in dialogue systems evaluated on the ConvEval benchmark. 13 claims were extracted from source…
Abstract: This report synthesises findings from 15 peer-reviewed papers addressing the following research question: Comprehensive comparison of frontier large language models on mathematical reasoning code generation and scientific knowledge v8. 18 claims were extracted from source literature; 0 were independently verified…
Abstract: This report synthesises findings from 13 peer-reviewed papers addressing the following research question: Which frontier language models achieve highest scores on GPQA Diamond Humanity Last Exam and difficult reasoning benchmarks v8. 10 claims were extracted from source literature; 10 were independently verified…
Abstract: This report synthesises findings from 4 peer-reviewed papers addressing the following research question: What are the state-of-the-art large language model results on reasoning benchmarks published recently v8. 7 claims were extracted from source literature; 7 were independently verified against retrieved documents.…
Abstract: This report synthesises findings from 9 peer-reviewed papers addressing the following research question: How does context length affect language model performance on multi-document reasoning and summarization v8. 10 claims were extracted from source literature; 10 were independently verified against retrieved…
Abstract: This report synthesises findings from 14 peer-reviewed papers addressing the following research question: What is the relationship between language model perplexity and downstream reasoning task performance v8. 9 claims were extracted from source literature; 9 were independently verified against retrieved documents.…
Abstract: This report synthesises findings from 11 peer-reviewed papers addressing the following research question: What is the effect of model size on language model performance on logical reasoning tasks v8. 8 claims were extracted from source literature; 8 were independently verified against retrieved documents. An…
Abstract: This report synthesises findings from 12 peer-reviewed papers addressing the following research question: How do language models compare to human experts on professional knowledge and science benchmarks v8. 9 claims were extracted from source literature; 9 were independently verified against retrieved documents. An…
Abstract: This report synthesises findings from 13 peer-reviewed papers addressing the following research question: How does synthetic training data improve language model performance on mathematical reasoning benchmarks v8. 9 claims were extracted from source literature; 9 were independently verified against retrieved…
Abstract: This report synthesises findings from 4 peer-reviewed papers addressing the following research question: What are the limitations of current language model evaluation benchmarks for measuring reasoning v8. 7 claims were extracted from source literature; 7 were independently verified against retrieved documents. An…
Abstract: This report synthesises findings from 14 peer-reviewed papers addressing the following research question: What prompting strategies maximize language model accuracy on graduate-level science questions v8. 12 claims were extracted from source literature; 8 were independently verified against retrieved documents. An…
Abstract: This report synthesises findings from 14 peer-reviewed papers addressing the following research question: How does extended thinking time affect language model accuracy on competition-level mathematics v8. 9 claims were extracted from source literature; 9 were independently verified against retrieved documents. An…
Abstract: This report synthesises findings from 14 peer-reviewed papers addressing the following research question: How does model quantization affect reasoning capability in large language models v8. 9 claims were extracted from source literature; 9 were independently verified against retrieved documents. An automated…
Abstract: This report synthesises findings from 2 peer-reviewed papers addressing the following research question: How do language models handle multi-hop reasoning chains in scientific question answering v8. 19 claims were extracted from source literature; 3 were independently verified against retrieved documents. An…
Abstract: This report synthesises findings from 8 peer-reviewed papers addressing the following research question: What is the comparative performance of open-source language models versus proprietary models on coding benchmarks v8. 0 claims were extracted from source literature; 0 were independently verified against retrieved…