SuperMemory-VQA provides 4,853 human-verified QA pairs from 52.9 hours of egocentric AI glasses recordings to benchmark AI systems on realistic long-horizon memory tasks including an unanswerable option.
hub
arXiv preprint arXiv:2411.13093 , year=
19 Pith papers cite this work, alongside 1 external citations. Polarity classification is still indexing.
19
Pith papers citing it
1
external citations · external index
hub tools
citation-role summary
background 2
baseline 1
citation-polarity summary
representative citing papers
OASIS organizes streaming video into hierarchical events and retrieves memory on-demand via intent-driven refinement to improve long-horizon accuracy and compositional reasoning with bounded token costs.
SVAgent improves long video question answering by constructing storylines via multi-agent collaboration and aligning cross-modal predictions for more robust, human-like reasoning.
MM-Mem distills video input through a hierarchical memory of sensory buffer, episodic stream, and symbolic schema, optimized by a semantic information bottleneck and SIB-GRPO, to achieve SOTA on long-horizon video benchmarks.
MAVIS introduces a multi-agent framework that parses videos into a structured semantic library and uses logic-aware debate among agents to retrieve relevant videos competitively without task-specific fine-tuning.
MemoryCard organizes long videos into self-contained topic-aware Memory Cards that improve long-video QA accuracy by up to 21.8% relative under fixed visual-token budgets.
GOPAgen proposes integrating video codec GOPs with a motion agent, GOP tree reasoning, structural memory, and motion vector database to improve efficiency and motion detail in agentic long-video VQA, reporting gains on MotionBench and EgoSchema.
OProver-32B achieves top Pass@32 scores on MiniF2F, ProverBench, and PutnamBench by combining continued pretraining with iterative agentic proving, retrieval, SFT on repairs, and RL on unresolved cases using a 6.86M-proof dataset.
LDDR proposes a linear DPP-based dynamic-resolution frame sampler that achieves 3x speedup and up to 2.5-point gains on video MLLM benchmarks by selecting non-redundant frames and allocating tokens accordingly.
Pro2Assist continuously tracks user progress in procedural tasks from AR-glasses sensors and delivers proactive step-aware assistance, outperforming baselines in step accuracy and timing.
HiCrew improves long-form video question answering on EgoSchema and NExT-QA via a hybrid tree for temporal topology, question-aware captioning, and adaptive multi-agent planning, with gains in temporal and causal reasoning.
W-RAC decouples extraction from semantic planning via structured units and LLM grouping to match traditional retrieval performance at roughly 10x lower LLM token cost.
OneClip-RAG enables MLLMs to handle long videos via one-shot clip retrieval and unified chunking-retrieval, delivering performance gains like matching GPT-5 level on MLVU with high efficiency on standard GPUs.
AdaQ is a training-free adaptive quasi-Gaussian sampling method for keyframe selection that improves long-video understanding in MLLMs and can outperform GPT-4o with 64 frames.
VISTA mines multi-level event semantics via visual prompts, knowledge-enhanced retrieval, and proposal integration to improve long-video event prediction over existing LVLMs.
MACF decouples agent perception budgets from overall video length using latent token collaboration to scale video understanding in MLLMs beyond current limits.
SceneBench shows VLMs sharply lose accuracy on scene-level long-video questions, and Scene-RAG only partially mitigates that forgetting (+2.50%).
Kwai Keye-VL-2.0-30B-A3B is a 30B MoE model with 3B active parameters using DSA adaptation and MOPD distillation that reports SOTA results on video understanding and agent benchmarks.
UrbanClipAtlas integrates RAG, taxonomy-aware extraction, and video grounding into a chat interface for retrieving and interpreting events in long urban videos from street intersections.
citing papers explorer
-
SuperMemory-VQA: An Egocentric Visual Question-Answering Benchmark for Long-Horizon Memory
SuperMemory-VQA provides 4,853 human-verified QA pairs from 52.9 hours of egocentric AI glasses recordings to benchmark AI systems on realistic long-horizon memory tasks including an unanswerable option.
-
OASIS: On-Demand Hierarchical Event Memory for Streaming Video Reasoning
OASIS organizes streaming video into hierarchical events and retrieves memory on-demand via intent-driven refinement to improve long-horizon accuracy and compositional reasoning with bounded token costs.
-
SVAgent: Storyline-Guided Long Video Understanding via Cross-Modal Multi-Agent Collaboration
SVAgent improves long video question answering by constructing storylines via multi-agent collaboration and aligning cross-modal predictions for more robust, human-like reasoning.
-
From Verbatim to Gist: Distilling Pyramidal Multimodal Memory via Semantic Information Bottleneck for Long-Horizon Video Agents
MM-Mem distills video input through a hierarchical memory of sensory buffer, episodic stream, and symbolic schema, optimized by a semantic information bottleneck and SIB-GRPO, to achieve SOTA on long-horizon video benchmarks.
-
MAVIS: Multi-Agent Video Retrieval via Structured Video Understanding
MAVIS introduces a multi-agent framework that parses videos into a structured semantic library and uses logic-aware debate among agents to retrieve relevant videos competitively without task-specific fine-tuning.
-
MemoryCard: Topic-Aware Multi-Modal Clue Compression for Long-Video Question Answering
MemoryCard organizes long videos into self-contained topic-aware Memory Cards that improve long-video QA accuracy by up to 21.8% relative under fixed visual-token budgets.
-
GOPAgen: Motion-Aware and Efficient Agentic Long-Video Understanding with Structural Memory and Hierarchical Reasoning
GOPAgen proposes integrating video codec GOPs with a motion agent, GOP tree reasoning, structural memory, and motion vector database to improve efficiency and motion detail in agentic long-video VQA, reporting gains on MotionBench and EgoSchema.
-
OProver: A Unified Framework for Agentic Formal Theorem Proving
OProver-32B achieves top Pass@32 scores on MiniF2F, ProverBench, and PutnamBench by combining continued pretraining with iterative agentic proving, retrieval, SFT on repairs, and RL on unresolved cases using a 6.86M-proof dataset.
-
LDDR: Linear-DPP-Based Dynamic-Resolution Frame Sampling for Video MLLMs
LDDR proposes a linear DPP-based dynamic-resolution frame sampler that achieves 3x speedup and up to 2.5-point gains on video MLLM benchmarks by selecting non-redundant frames and allocating tokens accordingly.
-
Pro$^2$Assist: Continuous Step-aware Proactive Assistance with Multi-modal Egocentric Perception for Long-horizon Procedural Tasks
Pro2Assist continuously tracks user progress in procedural tasks from AR-glasses sensors and delivers proactive step-aware assistance, outperforming baselines in step accuracy and timing.
-
HiCrew: Hierarchical Reasoning for Long-Form Video Understanding via Question-Aware Multi-Agent Collaboration
HiCrew improves long-form video question answering on EgoSchema and NExT-QA via a hybrid tree for temporal topology, question-aware captioning, and adaptive multi-agent planning, with gains in temporal and causal reasoning.
-
Web Retrieval-Aware Chunking (W-RAC) for Efficient and Cost-Effective Retrieval-Augmented Generation Systems
W-RAC decouples extraction from semantic planning via structured units and LLM grouping to match traditional retrieval performance at roughly 10x lower LLM token cost.
-
Towards Effective Long Video Understanding of Multimodal Large Language Models via One-shot Clip Retrieval
OneClip-RAG enables MLLMs to handle long videos via one-shot clip retrieval and unified chunking-retrieval, delivering performance gains like matching GPT-5 level on MLVU with high efficiency on standard GPUs.
-
Towards Fast and Effective Long Video Understanding of Multimodal Large Language Models via Adaptive Quasi-Gaussian Sampling
AdaQ is a training-free adaptive quasi-Gaussian sampling method for keyframe selection that improves long-video understanding in MLLMs and can outperform GPT-4o with 64 frames.
-
Towards Effective Long-Video Event Prediction via Multi-Level Event Semantics Mining
VISTA mines multi-level event semantics via visual prompts, knowledge-enhanced retrieval, and proposal integration to improve long-video event prediction over existing LVLMs.
-
Scaling Video Understanding via Compact Latent Multi-Agent Collaboration
MACF decouples agent perception budgets from overall video length using latent token collaboration to scale video understanding in MLLMs beyond current limits.
-
Seeing the Scene Matters: Revealing Forgetting in Video Understanding Models with a Scene-Aware Long-Video Benchmark
SceneBench shows VLMs sharply lose accuracy on scene-level long-video questions, and Scene-RAG only partially mitigates that forgetting (+2.50%).
-
Kwai Keye-VL-2.0 Technical Report
Kwai Keye-VL-2.0-30B-A3B is a 30B MoE model with 3B active parameters using DSA adaptation and MOPD distillation that reports SOTA results on video understanding and agent benchmarks.
-
UrbanClipAtlas: A Visual Analytics Framework for Event and Scene Retrieval in Urban Videos
UrbanClipAtlas integrates RAG, taxonomy-aware extraction, and video grounding into a chat interface for retrieving and interpreting events in long urban videos from street intersections.