MedHorizon benchmark reveals current multimodal LLMs achieve only 41.1% accuracy on long medical videos due to failures in sparse evidence retrieval and procedural reasoning.
Title resolution pending
9 Pith papers cite this work. Polarity classification is still indexing.
9
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 9roles
background 1polarities
background 1representative citing papers
Phoenix-bench shows agentic AI systems lose 37-58% resolved rate when moving from SWE-bench Verified to hardware tasks because bugs spread across parallel modules via signal flow, with testbench feedback lifting performance by 42-45% while file-level oracles add only 1.4%.
ProactBench measures LLM conversational proactivity in three phases using 198 multi-agent dialogues and finds recovery behavior hard to predict from existing benchmarks.
EvoPolicyGym is a new benchmark suite of 16 compact RL environments that evaluates autonomous policy evolution, with GPT-5.5 achieving the top aggregate rank and top-two performance on all tasks.
RDM trains one-step generators via MMD on large batches and multi-encoder representations, achieving SOTA SW_r14 of 1.30 on ImageNet and distilling FLUX.2 to one-step with gains on GenEval and PickScore.
Frontier LLMs miss dangerous actions in long coding agent transcripts 2-30 times more often after hundreds of thousands of benign tokens.
A deliberative council of Gemini agents using absence-based clinical rules achieves 0.382 F1 without fine-tuning and second place overall at 0.406 F1 on defense mechanism classification, with minority-class overrides adding 2.4pp.
Assertive linguistic features in training data increase LLMs' pro-animal-welfare reasoning while hedged and sensory-description features decrease it.
citing papers explorer
-
MedHorizon: Towards Long-context Medical Video Understanding in the Wild
MedHorizon benchmark reveals current multimodal LLMs achieve only 41.1% accuracy on long medical videos due to failures in sparse evidence retrieval and procedural reasoning.
-
Is Agentic AI Ready for Real-World Hardware Engineering? A Deep Dive with Phoenix-bench
Phoenix-bench shows agentic AI systems lose 37-58% resolved rate when moving from SWE-bench Verified to hardware tasks because bugs spread across parallel modules via signal flow, with testbench feedback lifting performance by 42-45% while file-level oracles add only 1.4%.
-
ProactBench: Beyond What The User Asked For
ProactBench measures LLM conversational proactivity in three phases using 198 multi-agent dialogues and finds recovery behavior hard to predict from existing benchmarks.
-
EvoPolicyGym: Evaluating Autonomous Policy Evolution in Interactive Environments
EvoPolicyGym is a new benchmark suite of 16 compact RL environments that evaluates autonomous policy evolution, with GPT-5.5 achieving the top aggregate rank and top-two performance on all tasks.
-
Representation Distribution Matching for One-Step Visual Generation
RDM trains one-step generators via MMD on large batches and multi-encoder representations, achieving SOTA SW_r14 of 1.30 on ImageNet and distilling FLUX.2 to one-step with gains on GenEval and PickScore.
-
Classifier Context Rot: Monitor Performance Degrades with Context Length
Frontier LLMs miss dangerous actions in long coding agent transcripts 2-30 times more often after hundreds of thousands of benign tokens.
-
UTS at PsyDefDetect: Multi-Agent Councils and Absence-Based Reasoning for Defense Mechanism Classification
A deliberative council of Gemini agents using absence-based clinical rules achieves 0.382 F1 without fine-tuning and second place overall at 0.406 F1 on defense mechanism classification, with minority-class overrides adding 2.4pp.
-
Assert, don't describe: Linguistic features that shift LLM reasoning about animal welfare
Assertive linguistic features in training data increase LLMs' pro-animal-welfare reasoning while hedged and sensory-description features decrease it.
- Repair the Amplifier, Not the Symptom: Stable World-Model Correction for Agent Rollouts