BrowseComp+
PulseAugur coverage of BrowseComp+ — every cluster mentioning BrowseComp+ across labs, papers, and developer communities, ranked by signal.
- used by Gotit.pub 70%
- instance of CatalyzeX 70%
- used by alphaXiv 70%
- used by ScienceCast 70%
- used by CatalyzeX 70%
- used by DagsHub 70%
- used by Grpo 70%
- instance of Gotit.pub 70%
- used by BrowseComp-ZH 70%
- instance of BrowseComp-ZH 70%
- instance of DeepSearchQA 70%
- instance of Generative Ai Interactive Agents 70%
6 day(s) with sentiment data
-
Mistral Large 4 released with 1T parameters, open weights, and 1M context · 1 source tracked
Mistral AI has officially released Mistral Large 4, also known as "Le Chonk," a 1.05 trillion parameter open-weight model. This model features 52 billion active parameters per token, a 1 million token context window, an…
-
Anthropic releases Claude 4 and expands global availability · 8 sources tracked
Anthropic has announced the release of Claude 4, marking a significant step in their AI development. The company is also expanding Claude's availability, making it accessible in Brazil and the European Union. Additional…
-
New LEAD-RL framework boosts multi-agent research efficiency
Researchers have developed LEAD-RL, a novel training framework for multi-agent deep research systems that improves coordination and efficiency. This approach uses contribution-aware credit assignment to better evaluate …
-
Meituan-longcat releases LongCat-2.5 multimodal agent model
Meituan-longcat has released LongCat-2.5, a multimodal Mixture-of-Experts model designed for long-horizon agentic tasks. This model boasts 1.6 trillion total parameters, with approximately 48 billion activated per token…
-
SenseTime launches free multimodal agent SenseNova 6.8 Flash
SenseTime has launched its new multimodal agent model, SenseNova 6.8 Flash, which is currently available for free. This model demonstrates strong capabilities in handling complex web tasks that have challenged other lea…
-
Nemotron 3.5 Bolt 74B-A9B 24GB: New Hybrid Vision-Language Model for Local AI
Researchers have introduced Nemotron 3.5 Bolt 74B-A9B 24GB, a novel hybrid vision-language model designed for efficient local AI applications. This model, derived from the Nemotron 3.5 Super VL 120B-A12B family, utilize…
-
Cerebras details trillion-parameter AI model deployment on wafer-scale chips
Cerebras has detailed its method for deploying trillion-parameter AI models on its WSE-3T wafer-scale chips. The company explained how it splits large models across multiple wafers, utilizing on-chip memory to store mod…
-
Exa 3.0 Search API claims benchmark lead over Perplexity, Brave
Exa Labs has launched Exa 3.0, an updated version of its web search API designed for AI agents. This release introduces Exa Lite, a more cost-effective search option, and a larger embedding model. Exa claims its new API…
-
FLINT architecture boosts retrieval model efficiency and scalability
Researchers have introduced FLINT (Fixed-budget Late INTeraction), a novel architecture designed to improve the scalability of multi-vector retrieval models. FLINT employs attention-based pooling to represent queries an…
-
New HybridResearcher agent bridges structured and unstructured data
Researchers have developed HybridResearcher, a 4B agent designed to integrate information from both structured and unstructured data sources. This agent was trained using a novel method of synthesizing cross-source reas…
-
New SEF method enhances AI agent training and benchmark performance
Researchers have developed a new method called Self-Reported Entity Familiarity (SEF) to improve the training of deep-search agents. This technique involves asking the teacher model if it is familiar with entities befor…
-
New REMORY network improves LLM context compaction for long-horizon agents
Researchers have developed REMORY, a neural memory network designed to enhance the context compaction capabilities of long-horizon agents. REMORY supplements a textual summary with soft memory tokens, allowing frozen la…
-
7B model ZGCM-1 prioritizes tool use and large context over memorization
Researchers from Zhongguancun Academy and Zhongguancun Institute of AI have developed ZGCM-1, a 7.39B parameter model that prioritizes tool use and a large context window over memorizing vast datasets. This approach all…
-
Open-source Iris search agent challenges closed-source rivals with advanced context management
AllSpark Research has launched Iris, an open-source search agent that challenges closed-source competitors. Iris utilizes a Mixture-of-Experts architecture and features a 256K context window, with models available under…
-
ByteDance's HarnessDev benchmark tests LLMs' ability to build agent code
Researchers from ByteDance Seed and other institutions have introduced HarnessDev, a new benchmark designed to evaluate an LLM's ability to create its own agent harnesses. Unlike traditional benchmarks that fix the harn…
-
OpenResearcher pipeline enables offline synthesis of AI research trajectories
Researchers have developed OpenResearcher, an open-source pipeline designed for synthesizing long-horizon research trajectories for training deep research agents. This pipeline operates offline, utilizing three explicit…
-
AllSpark Research unveils Iris, an open-weight web search agent
AllSpark Research has introduced Iris, an open-weight web search agent system designed to tackle complex, multi-hop questions that often stump current language models. Iris utilizes two models, Iris-mini and Iris-pro, p…
-
Iris search agents achieve SOTA open-source results on web benchmarks · 2 sources tracked
Researchers have developed two large-scale search agents, Iris-mini and Iris-pro, trained at 35B and 397B parameters respectively. These agents utilize a novel data pipeline and training methodology that combines superv…
-
New ICA framework improves AI agents' long-horizon information seeking
Researchers have developed a new framework called Information-Aware Credit Assignment (ICA) to improve reinforcement learning for agents that seek information over long horizons. ICA addresses the challenge of assigning…
-
Perplexity unveils local-first agent with PPLX 27B model
Perplexity has released new research detailing its "Portable Computer" agent, designed for local-first, private, and cost-effective work. This agent utilizes a post-trained PPLX 27B model, achieving high accuracy on kno…