Human Feedback and Eval Paper Explorer

Match reason: Keyword overlap 2/2 across title and protocol fields.

• Evaluations on real-world traces from the Azure LLM Inference Dataset and LMSYS-Chat-1M, serving Llama-3-70B on A100 GPUs, show that our approach reduces GPU-hours by 31-42%, corresponding to \2.86M annual savings at fleet scale, while…

Match reason: Keyword overlap 2/2 across title and protocol fields.

• Abstract shows limited direct human-feedback or evaluation-protocol detail; use as adjacent methodological context.

Match reason: Keyword overlap 2/2 across title and protocol fields.

• Across multiple advanced reasoning benchmarks, aPSF outperforms strong baselines including principle-aware optimizers, improving accuracy by up to +2.16 percentage points on average, and reduces optimization cost by 45--87% tokens on…

Match reason: Keyword overlap 2/2 across title and protocol fields.

• Abstract shows limited direct human-feedback or evaluation-protocol detail; use as adjacent methodological context.

Match reason: Keyword overlap 2/2 across title and protocol fields.

• We evaluate CoDE-Stop on diverse reasoning and science benchmarks across multiple models.

Match reason: Keyword overlap 2/2 across title and protocol fields.

• Abstract shows limited direct human-feedback or evaluation-protocol detail; use as adjacent methodological context.

Match reason: Keyword overlap 2/2 across title and protocol fields.

• Abstract shows limited direct human-feedback or evaluation-protocol detail; use as adjacent methodological context.

Match reason: Keyword overlap 2/2 across title and protocol fields.

• Recently, LLM-based agents have become increasingly popular across many applications, including complex sequential decision-making problems.
• In ReDAct, an agent is equipped with two LLMs: a small, cheap model used by default, and a large, more reliable but expensive model.

Match reason: Keyword overlap 2/2 across title and protocol fields.

• Across five model configurations, two families, and three benchmarks, we find that 52--88% of chain-of-thought tokens are produced after the answer is recoverable from a partial prefix.

Match reason: Keyword overlap 1/2 across title and protocol fields.

• Arabic-DeepSeek-R1 achieves the highest average score across the seven-benchmark OALL suite while establishing SOTA or near-SOTA, including dominant results on grammar-focused MadinahQA (surpassing both GPT-5.1 and the OALL leader by…
• Our results indicate that the combination of sparse MoE architecture, culturally-informed CoT distillation with explicit Arabic linguistic checks, and strategic bilingual data curation enables an open-source adapted model to systematically…

Match reason: Keyword overlap 1/2 across title and protocol fields.

• Extensive evaluations across three tasks demonstrate that our approach reduces prefilling FLOPs by 27.48\% while maintaining competitive accuracy.

Match reason: Keyword overlap 1/2 across title and protocol fields.

• Across both 1.5B and 4B model families, BCR reduces token usage by 15.8% to 62.6% while consistently maintaining or improving accuracy across five major mathematical benchmarks.

Match reason: Keyword overlap 1/2 across title and protocol fields.

• Abstract shows limited direct human-feedback or evaluation-protocol detail; use as adjacent methodological context.

Match reason: Keyword overlap 1/2 across title and protocol fields.

• Large Language Model (LLM) agents require persistent memory to maintain personalization, factual continuity, and long-horizon reasoning, yet standard context-window and retrieval-augmented generation (RAG) pipelines degrade over…
• Across benchmarks, MemMachine achieves strong accuracy-efficiency tradeoffs: on LoCoMo it reaches 0.9169 using gpt4.1-mini; on LongMemEvalS (ICLR 2025), a six-dimension ablation yields 93.0 percent accuracy, with retrieval-stage…

Match reason: Keyword overlap 1/2 across title and protocol fields.

• Evaluated across five benchmarks and two model scales, SRPO achieves both the rapid early improvement of SDPO and the long-horizon stability of GRPO.
• It consistently surpasses the peak performance of both baselines, raising the five-benchmark average on Qwen3-8B by 3.4% over GRPO and 6.3% over SDPO, while simultaneously yielding moderate response lengths and lowering per-step compute…

Match reason: Keyword overlap 1/2 across title and protocol fields.

• We design evaluation methods to detect position bias and token bias and training-free debiasing methods.

Match reason: Matched by broad semantic/index fallback.

• However, existing LLM-based approaches typically attempt to replace human experts directly without integrating effectively with existing SAST tools.
• We introduce Argus (Agentic and Retrieval-Augmented Guarding System), the first multi-agent framework designed specifically for vulnerability detection.

Match reason: Matched by broad semantic/index fallback.

• Abstract shows limited direct human-feedback or evaluation-protocol detail; use as adjacent methodological context.

Match reason: Matched by broad semantic/index fallback.

• Abstract shows limited direct human-feedback or evaluation-protocol detail; use as adjacent methodological context.

Match reason: Matched by broad semantic/index fallback.

• Crystalite achieves state-of-the-art results on crystal structure prediction benchmarks, and de novo generation performance, attaining the best S.U.N.