Skip to content

Commit c7de7f8

Browse files
authored
docs(paper-oct-14): add papers (#80)
add paper and update browsecomp config
1 parent 33aeecc commit c7de7f8

File tree

2 files changed

+42
-26
lines changed

2 files changed

+42
-26
lines changed

config/agent_browsecomp-en_mirothinker.yaml

Lines changed: 6 additions & 26 deletions
Original file line numberDiff line numberDiff line change
@@ -22,6 +22,11 @@ main_agent:
2222

2323
tool_config:
2424
- tool-reasoning
25+
- tool-searching
26+
- tool-image-video
27+
- tool-reading
28+
- tool-code
29+
- tool-audio
2530

2631
max_turns: 50 # Maximum number of turns for main agent execution
2732
max_tool_calls_per_turn: 10 # Maximum number of tool calls per turn
@@ -40,32 +45,7 @@ main_agent:
4045
chinese_context: "${oc.env:CHINESE_CONTEXT,false}"
4146

4247

43-
sub_agents:
44-
agent-worker:
45-
prompt_class: SubAgentWorkerPrompt
46-
llm:
47-
provider_class: "MiroThinkerSGLangClient"
48-
model_name: "DUMMY_MODEL_NAME"
49-
async_client: true
50-
temperature: 0.3
51-
top_p: 1.0
52-
min_p: 0.0
53-
top_k: -1
54-
max_tokens: 4096
55-
oai_mirothinker_api_key: "${oc.env:OAI_MIROTHINKER_API_KEY,dummy_key}"
56-
oai_mirothinker_base_url: "${oc.env:OAI_MIROTHINKER_BASE_URL,http://localhost:61005/v1}"
57-
keep_tool_result: -1
58-
oai_tool_thinking: false
59-
60-
tool_config:
61-
- tool-searching
62-
- tool-image-video
63-
- tool-reading
64-
- tool-code
65-
- tool-audio
66-
67-
max_turns: 50 # Maximum number of turns for main agent execution
68-
max_tool_calls_per_turn: 10 # Maximum number of tool calls per turn
48+
sub_agents: null
6949

7050

7151
# Can define some top-level or default parameters here

docs/mkdocs/docs/all_about_agents.md

Lines changed: 36 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -239,6 +239,34 @@ Welcome to our comprehensive resource collection for AI agents. This page curate
239239
**P073** - Auto-scaling Continuous Memory for GUI Agent
240240
- [:material-file-document: Paper](https://arxiv.org/abs/2510.09038)
241241

242+
**P074** - StoryBox: Collaborative Multi-Agent Simulation for Hybrid Bottom-Up Long-Form Story Generation Using Large Language Models
243+
- [:material-file-document: Paper](https://arxiv.org/abs/2510.11618)
244+
245+
**P075** - WebRouter: Query-specific Router via Variational Information Bottleneck for Cost-sensitive Web Agent
246+
- [:material-file-document: Paper](https://arxiv.org/abs/2510.11221)
247+
248+
**P076** - LLM×MapReduce-V3: Enabling Interactive In-Depth Survey Generation through a MCP-Driven Hierarchically Modular Agent System
249+
- [:material-file-document: Paper](https://arxiv.org/abs/2510.10890)
250+
251+
**P077** - BrowserAgent: Building Web Agents with Human-Inspired Web Browsing Actions
252+
- [:material-file-document: Paper](https://arxiv.org/abs/2510.10666)
253+
254+
**P078** - AGENTIQL: An Agent-Inspired Multi-Expert Framework for Text-to-SQL Generation
255+
- [:material-file-document: Paper](https://arxiv.org/abs/2510.10661)
256+
257+
**P079** - FML-bench: A Benchmark for Automatic ML Research Agents Highlighting the Importance of Exploration Breadth
258+
- [:material-file-document: Paper](https://arxiv.org/abs/2510.10472)
259+
260+
**P080** - MedAgentAudit: Diagnosing and Quantifying Collaborative Failure Modes in Medical Multi-Agent Systems
261+
- [:material-file-document: Paper](https://arxiv.org/abs/2510.10185)
262+
263+
**P081** - Can Tool-Integrated Reinforcement Learning Generalize Across Diverse Domains?
264+
- [:material-file-document: Paper](https://arxiv.org/abs/2510.11184)
265+
266+
**P082** - A Survey on Agentic Multimodal Large Language Models
267+
- [:material-file-document: Paper](https://arxiv.org/abs/2510.10991)
268+
269+
242270

243271
---
244272

@@ -361,6 +389,12 @@ Welcome to our comprehensive resource collection for AI agents. This page curate
361389
**E029** - DITING: A Multi-Agent Evaluation Framework for Benchmarking Web Novel Translation
362390
- [:material-file-document: Paper](https://arxiv.org/abs/2510.09116)
363391

392+
**E030** - When Agents Trade: Live Multi-Market Trading Benchmark for LLM Agents
393+
- [:material-file-document: Paper](https://arxiv.org/abs/2510.11695)
394+
395+
**E031** - A Comprehensive Survey on Benchmarks and Solutions in Software Engineering of LLM-Empowered Agentic System
396+
- [:material-file-document: Paper](https://arxiv.org/abs/2510.09721)
397+
364398

365399
---
366400

@@ -420,6 +454,8 @@ Welcome to our comprehensive resource collection for AI agents. This page curate
420454
**M017** - Mem-α: Learning Memory Construction via Reinforcement Learning
421455
- [:material-file-document: Paper](https://arxiv.org/abs/2509.25911)
422456

457+
**M018** - Preference-Aware Memory Update for Long-Term LLM Agents
458+
- [:material-file-document: Paper](https://arxiv.org/abs/2510.09720)
423459

424460
---
425461

0 commit comments

Comments
 (0)