claim

active

claim:atlas-combines-the-strengths-of-agentic-and-latent-reasoning-by-using-a-single-discrete-word-functional-token-that-serves-both-roles

ATLAS combines the strengths of agentic and latent reasoning by using a single discrete word (functional token) that serves both roles.

Core claim of the paper: a unified token bridges the gaps.

Source paper

extracted_from

ATLAS: Agentic or Latent Visual Reasoning? One Word is Enough for Both

Ziyu Guo · Rain Liu · Xinyan Chen · Pheng-Ann Heng

Neighborhood — ranked by edge-count

Papers (1)

paper

ATLAS: Agentic or Latent Visual Reasoning? One Word is Enough for Both
mentions

Communities (3)

community

Active inference & agent ecology
members_of
Free energy minimization, Markov blankets, trust gradients, and multi-agent rhythm/deferral frameworks
Hybrid agentic-latent reasoning architectures
members_of
Methods bridging external tool-use agents and internal hidden reasoning, examining trade-offs like context-switching latency and unified token representations.
Functional token unified reasoning
members_of
Single discrete word bridging agentic and latent visual reasoning in ATLAS framework

Frameworks (1)

framework

ATLAS Framework
cites
A framework where a single discrete word (functional token) serves both agentic operation and latent visual reasoning, requiring no visual supervision.

Concepts (1)

concept

Functional Token
cites
A discrete token in the vocabulary that represents a visual operation (e.g., <|Line|>, <|Shape|>, <|Text|>), generated via next-token prediction within autoregressive sequences.

Related by similarity (8)

cosine ≥ 0.65 · no typed edge

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

ATLAS: Agentic or Latent Visual Reasoningframework0.865
Core framework proposing discrete functional tokens as a unified solution for visual reasoning in VLMs, bridging agentic and latent approaches.
One word is enough for both agentic and latent visual reasoning.claim0.798
The paper's thesis from the title.
One word is enough for both agentic operation and latent reasoning unitconcept0.774
Core idea of ATLAS: a single discrete token serves dual purpose of operation specification and latent reasoning.
Recent alternatives include agentic reasoning through code or tool calls, and latent reasoning with learnable hidden embeddings.claim0.747
Statement of existing alternatives to direct generation.
Five functional tokens can generalize across 40+ diverse visual reasoning taskshypothesis0.744
ATLAS hypothesis that a compact set of high-level functional tokens (Manip, Shape, Line, Arrow, Text) suffices for multi-domain visual reasoning.
Delegation along multiple axes is not meant to imply any kind of concurrent or nondeterministic mechanism.claim0.742
Clarification that the multi-axis delegation is sequential and deterministic.
Evaluating agentic competence requires benchmarks that test the joint deployment of multiple capabilities in multi-agent environments with conflicting incentives, uncertainty, and economic dynamics.quote0.737
Motivational statement for the benchmark design philosophy.
Each functional token is associated with an internalized visual operation, yet requires no visual supervision and remains a standard token in the tokenizer vocabulary.claim0.731
Describes the properties of the functional token.