On Resolving Ambiguous Anaphoric Expressions in Imperative Discourse
Authors: Vasanth Sarathy, Matthias Scheutz6957-6964
AAAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In this paper, we introduce, with examples, a general class of situated anaphora resolution problems, propose a proof-of-concept system for disambiguating situated pronouns, and discuss some general types of reasoning that might be needed. The results from computing the fused uncertainties for each of discourse examples (3), (4) and (5) are shown in Table 1. |
| Researcher Affiliation | Academia | Vasanth Sarathy Department of Computer Science Tufts University 200 Boston Ave. Medford, MA 02155 vasanth.sarathy@tufts.edu Matthias Scheutz Department of Computer Science Tufts University 200 Boston Ave. Medford, MA 02155 matthias.scheutz@tufts.edu |
| Pseudocode | Yes | The paper includes structured code-like blocks, specifically Answer Set Programming (ASP) rules under sections like "Plausibility Microtheory", "Normative Microtheory", and "Speaker Intent Microtheory". For example: "#program base. % Percept types is(obj1,object). ... %GENERATE { occ(A,t) : action(A) } = 1." These serve as algorithmic descriptions. |
| Open Source Code | No | The paper does not provide concrete access to its own source code. It only mentions a link to a third-party neural coreference system: "Figure 3: Incorrect resolution of a current neural coreference system (Clark and Manning 2016) on the example shown in Figure 1. (Implementation: https://huggingface.co/coref/)". |
| Open Datasets | No | The paper does not provide concrete access information for a publicly available or open dataset used in its experiments. It uses specific examples (3), (4), and (5) to demonstrate its proof-of-concept system rather than evaluating on a formal, accessible dataset. |
| Dataset Splits | No | The paper does not provide specific dataset split information. It evaluates its proof-of-concept system on a few illustrative examples (3), (4), and (5) rather than a formal dataset with defined training, validation, and test splits. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., GPU/CPU models, memory amounts) used for running its experiments. It only discusses the conceptual and software aspects of its system. |
| Software Dependencies | No | The paper mentions using "ASP implementations in clingo and iclingo" but does not provide specific version numbers for these software dependencies or any other libraries or frameworks. |
| Experiment Setup | Yes | This microtheory is essentially a planning microtheory defined over a short time horizon of five steps. The #program check(t) operation provides a termination of the program, which in this case is after five steps, set by the maxlength constant. #const maxlength=5. |