Proof pending. Core topic summary fields are still materializing.
Current research in AI focuses on enhancing the capabilities of large language models (LLMs) to improve their generalization and alignment with human values. Studies reveal limitations in LLMs' ability to generalize periodicity and their lack of a coherent Theory of Mind, which affects their social reasoning. New normalization techniques and independent modules are being developed to stabilize learning and ensure consistent value guidance. Additionally, advancements in latent reasoning and controllable information production aim to refine how models learn from human feedback and optimize their performance. These developments are crucial for builders looking to create more reliable and effective AI systems that can operate in diverse and complex environments.
Topic-specific paper and score movement from the daily diff ledger.
Large language models (LLMs) based on the Transformer have demonstrated strong performance across diverse tasks. However, current models still exhibit substantial limitations in out-of-distribution (O...
The normalization of query and key vectors is an essential part of the Transformer architecture. It ensures that learning is stable regardless of the scale of these vectors. Some normalization approac...
Do Large Language Models (LLMs) possess a Theory of Mind (ToM)? Research into this question has focused on evaluating LLMs against benchmarks and found success across a range of social tasks. However,...
We investigate whether \emph{LLM-based agents} can develop task-oriented communication protocols that differ from standard natural language in collaborative reasoning tasks. Our focus is on two core p...
We introduce the MiniMax-M2 series, a family of Mixture-of-Experts language models built around the principle that mini activations can unleash maximum real-world intelligence. The flagship M2 contain...
Intrinsic Motivation (IM) is a paradigm for generating intelligent behavior without external utilities. The existing information-theoretic methods for IM are predominantly based on information transmi...
Learning from human feedback typically relies on preference optimization that constrains policy updates through token-level regularization. However, preference optimization for language models is part...
We study two recurring phenomena in Transformer language models: massive activations, in which a small number of tokens exhibit extreme outliers in a few channels, and attention sinks, in which certai...
Aligning large language models (LLMs) with human values typically relies on post-training or inference-time steering that directly manipulates the backbone's parameters or representation space. Howeve...
Transformers excel at in-context retrieval but suffer from quadratic complexity with sequence length, while State Space Models (SSMs) offer efficient linear-time processing but have limited retrieval ...
Freshness
Canonical route: /topics
Agent Handoff
Canonical ID ai-research | Route /topic/ai-research
REST example
curl https://sciencetostartup.com/api/v1/agent-handoff/topic/ai-researchMCP example
{
"tool": "search_papers",
"arguments": {
"query": "AI Research",
"cluster": "AI Research"
}
}source_context
{
"surface": "topic",
"mode": "topic",
"query": "AI Research",
"normalized_query": "ai-research",
"route": "/topic/ai-research",
"paper_ref": null,
"topic_slug": "ai-research",
"benchmark_ref": null,
"dataset_ref": null
}Use This Via API or MCP
Topic pages bundle paper counts, viability trends, author concentration, and top questions into one canonical surface your agents can reference before they open Signal Canvas or create a workspace.